RIVolutionising Voluntary Disclosure

of 95

RIVolutionising Voluntary Disclosure: A Conceptual Study on the Value Relevance of Voluntary

Disclosure for Listed Emerging Growth Companies

Martina Fernanda Adriana Policastro* Maximilian Matheis°

Master Thesis in Accounting & Financial Management Stockholm School of Economics

December 2018

Abstract This study uses a mixed method approach to analyse how voluntary disclosure contributes to conceptually explain the valuation gap between fundamental accounting numbers and the stock market price for Listed Emerging Growth Companies (LEGCs). This is done by taking a deduced valuation model perspective using the Forecast Issues underlying the Residual Income Valuation model as a theoretical lens to analyse voluntary disclosure (Gray & Skogsvik, 2004). Although the quantitative analyses do not yield insightful results by themselves, the qualitative analysis allows to resolve the forecasting of the expected future book returns on owners’ equity (Forecast Issue 1) and the forecasting of the expected relative goodwill/ badwill of owners’ equity at the horizon point in time (Forecast Issue 2) on a short to medium term. This is highlighted by three key recurring themes that LEGCs disclose to explain their future value generation potential: Operating Leverage, Financial Myopia, and Supplemental Performance Reporting. This thesis is meant to facilitate further research on the topic by implementing a new approach to voluntary disclosure analysis, and to create interest in the research area of an upcoming type of firms, LEGCs.

Supervisor: Kenth Skogsvik, Professor, Department of Accounting Keywords: Voluntary disclosure, Residual income valuation, Value relevance, Listed

Emerging Growth Companies __________________

* [email protected] ° [email protected]

of 95

Acknowledgements We want to express our sincerest and warmest gratitude to our supervisor Kenth

Skogsvik for his valuable guidance, crucial insights and fruitful

advice in writing this thesis.

Stockholm, December 2018

of 95

Table of Contents 1 Introduction .......................................................................................... 6

2 Literature Review ................................................................................. 8

2.1 Voluntary Disclosure & Information Asymmetry ........................................... 8

2.2 Voluntary Disclosure & Value Relevance ..................................................... 11

2.2.1 Stock Liquidity ..................................................................................... 12

2.2.2 Cost of Capital ..................................................................................... 13

2.2.3 Method Shortcomings in Prior Literature ............................................. 15

2.3 Disclosure Environment for LEGCs ............................................................. 15

2.3.1 Current Trends in Disclosure ................................................................ 15

2.3.2 Disclosure & LEGCs ............................................................................. 16

3 The Residual Income Valuation Set-up ............................................... 19

3.1 Model Specification ...................................................................................... 19

3.2 Prediction Problems & The Need for More Information ............................... 21

3.2.1 Forecast Issue 1 .................................................................................... 22

3.2.2 Forecast Issue 2 .................................................................................... 23

3.2.3 Forecast Issue 3 .................................................................................... 25

3.2.4 Forecast Issue 4 .................................................................................... 25

4 Scope & Methodology ......................................................................... 27

4.1 Disclosure Scope .......................................................................................... 27

4.2 Data Collection ............................................................................................ 28

4.3 Approach to Analysis ................................................................................... 30

4.3.1 Quantitative Analysis ........................................................................... 30

4.3.2 Qualitative Analysis ............................................................................. 32

4.4 Data Legitimation ........................................................................................ 33

5 Sample Collection ............................................................................... 34

5.1 Collection of Main Sample ........................................................................... 34

5.2 Collection of Control Sample ....................................................................... 35

of 95

6 Data Description & Analysis ............................................................... 37

6.1 Quantitative Analysis .................................................................................. 37

6.1.1 Recorded Theme Distributions & Analysis ............................................ 37

6.1.2 Attribute Combinations & Sub-topic Level Analysis ............................. 41

6.1.3 Attribute Combinations & Node Level Analysis.................................... 45

6.2 Qualitative Analysis..................................................................................... 50

6.2.1 Regarding Forecast Issue 1 ................................................................... 50




6.3 Credibility ................................................................................................... 65

7 Discussion & Limitations .................................................................... 69

7.1 Voluntary Disclosure – A Solution to the Forecast Issues? ........................... 69

7.2 Non-Disclosure – ‘Window Dressing’ or Too Sensitive to Disclose? .............. 71

7.3 Limitations .................................................................................................. 74

8 Conclusion & Implications .................................................................. 75

8.1 Conclusion ................................................................................................... 75

8.2 Contributions to Literature .......................................................................... 76

8.3 Areas of Further Research ........................................................................... 77

9 References........................................................................................... 79

10 Appendices ......................................................................................... 85

10.1 Appendix A – EGC & SME Definition......................................................... 85

10.2 Appendix B – Table 7 .................................................................................. 87

10.3 Appendix C – Node Definitions .................................................................... 88

10.4 Appendix D – Table 8 ................................................................................. 93

10.5 Appendix E – Figure 8 & 9 .......................................................................... 94

of 95

List of Figures Figure 1 – Conceptual Illustration Book Returns on Owners’ Equity 22

Figure 2 – Conceptual Illustration q-Value Decomposition 24

Figure 3 – Example of Filled-Out Codification Matrix 31

Figure 4 – Conceptual Illustration of Gioia Methodology 32

Figure 5 – Gioia Analysis Summary Forecast Issue 1 (Main Sample) 51

Figure 6 – Gioia Analysis Summary Forecast Issue 2 (Main Sample) 58

Figure 7 – Gioia Analysis Summary Forecast Issue 3 & 4 (Both Samples) 63

Figure 8 – Gioia Analysis Summary Forecast Issue 1 (Control Sample) 94

Figure 9 – Gioia Analysis Summary Forecast Issue 2 (Control Sample) 95

List of Tables Table 1 – Sample Overview 36

Table 2 – Descriptive Statistics of Recorded Themes 38

Table 3 – Number of Recorded Themes in Specific Nodes 40

Table 4 – Distribution of Attribute Combinations within Sub-topics 43

Table 5 – Distribution of Attribute Combinations within Additional

Financial Data Nodes 46

Table 6 – Distribution of Attribute Combinations within Business Model Nodes 48

Table 7 – Document Count 87

Table 8 – Recorded Theme Distribution in Accounting Standard Clusters 93

of 95

1 Introduction

In 2017, 83% of all US IPOs in the technology sector were done by companies which

were loss-making in the 12 months prior to their listing (Ritter, 2018). While this trend

has been ongoing for the last seven years, it is even more remarkable that these firms

were able to generate, on average, a higher stock return than the S&P 500 index (36%

vs. 9% stock appreciation) since their IPO date (Driebusch & Farrell, 2018), despite their

negative trailing net income. This trend highlights the emergence of a new type of firms

which, regardless of the historical financial performance, conveys to the market the

promise of a sustained, above average growth along with a strong generation of value in

the future.

However, looking at it from a fundamental accounting perspective, the question can be

raised of how to explain the market value of this type of firms. Building on the

complexities mentioned by Damodaran (2009), it appears very challenging to value young

and unprofitable growth companies given their early business life-cycle stage, their

business models, which are not properly captured by the accounting mould, and

ultimately their substantial truncation risk. Using the mandatory information coming

from the reported financial disclosure as a starting point to reverse-engineer the value of

these firms, it becomes apparent that this type of information has not been developed to

explain the company’s value generation potential implicitly captured in the market price,

leaving a value gap between the reported mandatory numbers and the market value.

This suggests that these unprofitable companies could provide additional information to

shed more light on their future value generation potential, i.e. they could be engaging in

voluntary disclosure practices.

This thesis differentiates itself from prior literature by conceptually approaching the

topic of value relevance of voluntary disclosure from a deduced valuation model

perspective. This can be perceived as valuable, as a deduced valuation model approach

allows to make a meaningful assessment of the relationship between accounting numbers

and the stock market price and at the same time it simplifies prediction problems within

the modelling exercise. Moreover, further insights are generated by investigating this

upcoming type of firms, named Listed Emerging Growth Companies (LEGC), for the

of 95

remainder of this thesis. The term is used to define companies with the following shared

characteristics: a LEGC is listed, has been trading with liquid and free-floating shares

for more than one year, operates in an emerging industry as defined by PwC (2012),

predominantly belongs to the technology sector, and has a three-year revenue CAGR of

at least 10%.

Thus, this thesis is designed to answer the following research question:

How does voluntary disclosure contribute to conceptually explain the valuation gap

between fundamental accounting numbers and the stock market price of LEGCs within

a Residual Income Valuation set-up?

The research question is investigated by applying a mixed method approach starting off

from a high-level quantitative Thematic Content Analysis to a more thorough

quantitative Attribute Analysis as suggested by Beattie et al. (2004) and finishing off

with a qualitative analysis in line with the Gioia Methodology (Gioia et al., 2013), the

most granular way of analysing the data. The main sample, consisting of 25 LEGCs, is

controlled for by comparing its disclosure practice with that of a control sample

consisting of 10 ‘optimal-but-profitable twin’ LEGCs. The samples are compiled using a

purposive sampling method. A total of 117 financial disclosure documents in the main

sample and 47 in the control sample is codified and analysed.

This thesis contributes to the topic of value relevance of voluntary disclosure by

providing insights on how LEGCs address their value generation potential in financial

disclosure documents. The three recurring emerging themes are Operating Leverage,

Financial Myopia and Supplemental Performance Reporting, which contribute to explain

the value gap between fundamental accounting numbers and market value on a short to

medium term. Given the limited scope of the thesis, the purpose is to facilitate further

research on the topic and to create interest in an upcoming type of firms: LEGCs.

In order to investigate the outlined research question, the rest of this thesis is divided

into seven major sections. The first section provides a literature review of voluntary

disclosure and its value relevance. This is followed by the introduction of the deduced

of 95

valuation model in which the thesis is embedded, namely the Residual Income Valuation

model. The third part broaches the issue of the research scope and the underlying

methodology. Within the fourth part, the collection of the main and the control sample

is presented. Section five is divided into two sub-sections: the first dealing with the

quantitative analysis, followed by the second presenting the qualitative analysis. Section

six discusses the findings by integrating them into a broader debate about strategic

considerations in voluntary disclosure. To bring the thesis to a close, the seventh and

final section summarises the key findings and their contribution to existing literature

while at the same time outlining areas for potential future research.

2 Literature Review

To understand whether voluntary disclosure can help to conceptually explain the

valuation gap between underlying fundamental accounting numbers and the market price

for LEGCs such as Snap, Workday, Shopify or Delivery Hero, at first a review of the

existing literature on disclosure is conducted. Section 2.1 serves two purposes: first, it

distinguishes between mandatory and voluntary disclosure and second, as suggested by

Verrecchia (2001), it introduces the concept of information asymmetry as a starting point

to discuss the relevance of disclosure in a capital market and valuation setting. Section

2.2 then reviews two distinct aspects in which voluntary disclosure has shown to have

an economic impact, highlighting its value relevance for both management and external

users of reports such as investors and analysts. Section 2.3 presents current trends in the

reporting environment, which in turn leads to a questioning of the applicability of the

existing literature findings to LEGCs.

2.1 Voluntary Disclosure & Information Asymmetry

Analysing the capital market implications of information for investors, lenders and other

parties, inevitably leads to the traditional economic perspective whereby capital

allocation within markets is expected to be improved, since investment decisions by

shareholders, along with an assessment of risk-adjusted returns, are more profound

(Healy & Palepu, 2001). These involved parties require information not only to value the

future net cash flows that could derive from their invested resources but also to assess

whether management is efficiently and effectively using resources, i.e. fulfils its

of 95

stewardship function (IASB, 2018; FASB, 2018a), which are both key factors for their

valuation exercise and resource allocation process. Using information asymmetry as a

starting point for research in the area of voluntary disclosure (Verrecchia, 2001) and

combining it with the ‘lemons problem’ firstly introduced by Akerlof (1970), it could be

argued that, relative to the information available to management, investors and other

external parties have access to less information about future cash flows and stewardship,

which then translates into less efficient resource allocation decisions due to over-

(under- ) valuation of bad (good) companies.

The approach that has been widely accepted in order to reduce the information

asymmetry between investors and management is to enforce financial reporting (Healy

& Palepu, 2001). Financial disclosure is nowadays regulated by specific bodies (e.g.

Security and Exchange Commission in the US and Swedish Financial Reporting Board

in Sweden) and requirements are drafted by dedicated boards, such as the FASB and

the IASB, in order to ensure the availability of uniform and comparable information of

entities to investors. The underlying purpose of financial reporting, under both IFRS and

U.S.-GAAP, is to provide the necessary relevant and trustworthy information to the

primary users of financial statements, thereby supporting them to make educated

economic decisions (IASB, 2018; FASB, 2018a). Taking this into consideration, it can be

claimed that financial reporting frameworks are valuation-centric. In practice, financial

analysts’ primary source of information for their valuation exercise are essentially

financial reports. Besides the importance of financial reports for the capital market and

its intermediating parties, accounting data has been widely used within academia as a

source of information for research in accounting and financial analysis (Healy & Palepu,

2001).

Nevertheless, Healy & Palepu (1993) highlight three conditions under which financial

reporting is an imperfect communication channel between management and third parties:

information asymmetry, unaligned incentives of shareholders and management, as well

as imperfect accounting and audit rules. The IASB and the FASB themselves recognize

the third condition as true, both explicitly stating within their conceptual frameworks

that “general purpose financial reports do not and cannot provide all of the information

that existing and potential investors, lenders and other creditors need. Those users need

of 95

to consider pertinent information from other sources” (IASB, 2018, 1.6; FASB, 2018a,

OB.6), however without elaborating on what other sources they should look into.

A common approach established by management to overcome financial reporting

imperfection and increase the effectiveness of its communication with stakeholders is to

supplement mandatory with voluntary disclosure. Association-based studies, defined by

Verrecchia (2001, p. 97) as “work[s] that stud[y] the effect of exogenous disclosure on the

cumulative change or disruption in investors’ individual actions, primarily through the

behaviour of asset equilibrium prices and trading volume”, find voluntary disclosure to

be economically relevant for companies. Although, theoretical models and empirical

findings are discussed in detail within the subsequent section 2.2, at this point it is

important to note the relevance of association-based studies, since they provide a strong

argumentation for the consequences of companies’ management engaging in voluntary

disclosure practices. More recently, research statistically testing causality between

voluntary disclosure and firm value (e.g. Balakrishnan et al., 2014), indicates that a firm

can lower its cost of capital using voluntary disclosure due to a higher stock liquidity.

As this new area remains rather untapped when it comes to empirical findings, it is not

extensively addressed within the scope of this thesis.

A review of discretionary-based research, i.e. studies that examine how firms decide

which information to disclose, suggests that there are costs related to management

participating in voluntary disclosure (Verrecchia, 2001). Graham et al. (2005) conduct a

survey which highlights five major costs influencing the degree to which management

undertakes voluntary disclosure: 1. commitment costs, not committing to disclosure

precedents which management will not be able to maintain in the following periods; 2.

litigation costs, providing less information to reduce the possibility of lawsuits due to

incorrect information or stock volatility; 3. proprietary costs, not disclosing sensitive

information to avoid repercussions on a company’s competitive advantage; 4. agency

costs, not engaging in disclosure which could damage the agents reputation and future

career by not meeting the set expectations, and 5. political costs, not disclosing

information because it could bring unwanted attention from regulators.

of 95

It is also important to acknowledge the role of impression management (Leary &

Kowalski, 1990) in financial disclosure. As an example, studies have argued that

managers manipulate financial disclosure and use it opportunistically to influence the

share price (Adelberg, 1979; Rutherford, 2003; Courtis, 2004). The different manipulation

strategies that have been identified in research include, for example, Visual and

Structural Manipulation, Rhetorical Manipulation, and Choice of Earnings Number

(Merkl-Davies & Brennan, 2007). Evidence has shown that both financial (McGuire et

al., 1990) and non-financial (Black et al., 2000) reputation has significant value-relevance,

which could further induce management to engage in impression management practices

to ensure that the company retains positive reputation on the market. This is particularly

true for growth companies given that their stocks have asymmetrically large negative

price responses to negative earnings surprises (Skinner & Sloan, 2002).

To conclude, research on disclosure can be motivated by coming from the starting point

of information asymmetry. The general approach to close this information gap,

mandatory financial reporting, is perceived as being imperfect and is therefore often

supplemented by voluntary disclosure. Although studies have shown that voluntary

disclosure is value relevant, as it is described in the following section, there are both costs

and opportunistic behaviours which could hinder voluntary disclosure practices.

2.2 Voluntary Disclosure & Value Relevance

After having identified the role that voluntary disclosure plays in closing the information

asymmetry gap, it is necessary to get a deeper understanding of its economic implications

in a valuation setting. Looking into the association-based literature previously done on

financial reporting and disclosure within the capital market economy, three major aspects

resulting in stock market effects are identified: improved stock liquidity, reduced cost of

capital and increased information intermediation (Healy & Palepu, 2001). Core papers

with their respective findings, as well as contrasting views, are outlined in the following

sub-sections for the first two aspects, whereas information intermediation (see Lang &

Lundholm, 1993, 1996; Francis et al., 1997) is not deemed to be highly relevant for the

scope of this thesis given that the companies studied already have a sufficient analyst

following.

of 95

2.2.1 Stock Liquidity

Diamond and Verrecchia’s (1991) pivotal paper about the positive association between

additional disclosure and stock liquidity introduces a liquidity model with one firm and

two (ex-ante identical) larger institutional traders at three different points in time. It

essentially claims that disclosure improves future liquidity of a firm’s security resulting

from a reduction in information asymmetry, which induces larger institutional investors

to increase the competition with market makers. This, in turn, reduces the volatility of

upcoming order imbalances and ultimately causes market makers to leave.

Although Diamond and Verrecchia’s (1991) findings are developed in a theoretical

framework based on a set of assumptions, Healy et al. (1999) are able to further support

this by applying a time-series approach on a sample of 595 firms in 23 industries, thereby

revealing that expanded voluntary disclosure is followed by improved stock performance

and increased institutional ownership, analyst following and stock liquidity. These results

also hold when controlling for previously established explanatory variables such as

earnings performance, size and risk. Voluntary disclosure and its quality are hereby

measured using the Association of Investment Management and Research Corporate

Information Committee Reports (AIMR Reports), where subcommittees of industry-

specific financial analysts conduct surveys on an annual basis in order to rate companies’

disclosure practices as well as their improvements.

While there is a wide range of studies which explain the positive impact of voluntary

disclosure on stock liquidity, potential conflicts of interest between managers and outside

owners, due to a reduction in shareholder value, may arise since additional information

might not only reveal valuable insights for competitors, but also increase legal costs for

firms (e.g. Wagenhofer, 1990; Francis et al., 1994). Elaborating on this, it could be

claimed that voluntary disclosure by a company’s management is a very strategic process

which is not necessarily closing information asymmetries on the market, since it is inter

alia dependent on factors like proprietary cost or investor clientele.

More recent research by Schoenfeld (2017) therefore looks closer into index funds in order

to develop an empirical model independent from strategic disclosure motives given that

index funds trade primarily for non-strategic reasons. Within his sample of 368 new

of 95

entrant firms within the S&P 500 between 1996-2010, Schoenfeld (2017) derives that

voluntary disclosure increases with the level of index fund ownership, whereby the rise

in disclosure positively affects stock liquidity. Taking into special consideration the

aspect of index funds trading for non-strategic reasons, it is derived that they

unambiguously prefer and seem to demand higher disclosure to increase stock liquidity

and stock market efficiency. Consequently, Schoenfeld (2017) helps to understand that

disclosing certain private information to the market does not seem to depend on the

existence of an investor clientele with strategic trading motives (i.e. actively entering

long and short positions), thereby at least partially mitigating the concerns raised in the

previous paragraph regarding strategic disclosure behaviour.

Summarising the review on voluntary disclosure and stock liquidity, it can be claimed

that, despite some contrasting views with regard to shareholder value reduction (e.g.

legal costs or proprietary costs), there is a positive association between voluntarily

disclosed information and stock market liquidity.

2.2.2 Cost of Capital

While Diamond and Verrecchia’ s (1991) theoretical model already discusses the positive

impact of disclosure by lowering a company’s cost of capital due to higher future

liquidity, one of the first empirical evidences, provided by Botosan (1997), sheds even

more light on this topic. By regressing firm-specific estimates of equity cost of capital on

market beta, firm size and a self-developed measure of disclosure level for 122

manufacturing firms in the financial year 1990, she reveals that the more comprehensive

voluntary disclosure for firms with low analyst following is, the lower is their equity cost

of capital. Companies with high analyst following, however, do not indicate an

association between Botosan’s measure of disclosure level and the cost of equity, which

may be explained by the analysts having a crucial role in the financial communication

process. Brown et al. (2004), in addition, look specifically into conference calls as a

medium of voluntary disclosure and their impact on information asymmetry. Although

they are not explicitly testing the impact of conference calls on a company’s cost of

capital, their line of argumentation is that once information asymmetry is lowered, a

firm’s cost of capital decreases since the return premium, which compensates for the risk

of trading with privately informed investors, is lowered. By measuring information

of 95

asymmetry using the Probability of Informed Trade (PIN), they find strong cross-

sectional and time-series evidence that a firm’s conference call activity is negatively

related to the level of information asymmetry, ultimately lowering the firms’ cost of

capital. These findings, however, are only significant for companies which have a policy

of periodic conference calls.

A contrasting dimension, also with respect to the common earnings quality of companies

investigated for the purpose of this thesis, is outlined by Francis et al. (2008). By looking

into 677 firms’ annual reports and 10-K filings for 2001, they find evidence that although

greater (less) voluntary disclosure is associated with lower (higher) cost of capital

(unconditional on other factors), the disclosure effect on cost of capital is substantially

neglected or may even disappear once they control for earnings quality. The results turn

out to be robust when testing for potential differences in voluntary disclosure practice

across industries, estimation procedures (e.g. parametric or non-parametric) as well as

alternative measure for both earnings quality (e.g. accruals quality, earnings variability

or absolute value of abnormal accruals) and cost of capital (e.g. realized returns or debt

ratings). Potential endogeneity issues of firms’ whose disclosure choices determine the

earnings quality are diminished by Francis et al. (2008) since they reveal that voluntary

disclosure is associated with innate (i.e. quality under a firm’s production function and

business model) instead of discretionary earnings quality (i.e. quality under more

immediate management control).

What is hereby also interesting to note is that Francis et al. (2008) results derived from

their self-constructed index for annual filings behave differently from the outcome based

on proxies that are linked to management forecasts, conference calls or press releases. In

fact, management forecasts and conference calls have a positive association with cost of

equity, i.e. engaging in these forecasting activity leads to a higher cost of equity, whereas

no insightful relation could be found for press releases. While Francis et al. (2008) does

not provide an explicit explanation for the differing results across various voluntary

disclosure proxies, the authors raise the point that each proxy is characterized by

qualitatively different content as well as different intentions for disclosure by the

companies’ management, representing a rich area for further research. Although Francis

of 95

et al. (2008) results question the relevance of voluntary disclosure impact on cost of

equity, the majority of the research in this area is in line with Botosan’s (1997) findings.

2.2.3 Method Shortcomings in Prior Literature

A key concern that has to be raised when it comes to the association-based literature is

the way voluntary disclosure variables are constructed. Most of the research is conducted

using readily available analyst ratings on disclosure such as the AIMR or FAF Reports.

These proxies, however, were discontinued in 1997 (looking at the annual report of 1995),

thereby partially explaining why relevant literature fades away by the end of the 1990s.

Alternative solutions that are applied by other studies are the use of proxies (e.g.

management forecasts) or self-constructed measures (e.g. Botosan, 1997). Since these

data collection methods are prone to endogeneity issues, e.g. biases in selection or

analysts’ behaviour, and are usually overly aggregated, e.g. binary classification systems,

studies with these noisy measures of disclosure should be analysed carefully (Healy &

Palepu, 2001).

To summarise, the review of previous disclosure literature shows that mandatory

financial reporting is serving its purpose of bridging information gaps between

management and external parties, although it does so in an imperfect manner, justifying

the need for additional voluntary disclosure. Overall, management will have to find the

optimal balance when it comes to how much and what to voluntarily disclose in order to

maximise the potential economic gains and minimise the potential costs related to

disclosure (Verrecchia, 2001) while at the same time successfully providing the

information necessary to primary users of reports to make better resource allocation

decision and more accurate valuation exercises.

2.3 Disclosure Environment for LEGCs

2.3.1 Current Trends in Disclosure

Voluntary disclosure is not something companies are solely driving by themselves.

Indeed, in recent years, investors and regulatory bodies together have been requesting

for more information to be disclosed by companies. This includes all types of information

from financial KPIs (Deloitte, 2017) to corporate social responsibility or carbon emissions

of 95

(Michelon et al., 2015; KPMG, 2017a). Some information has become regulated and

mandatory, such as anti-corruption and bribery policies (EU Directive 95, 2014), albeit

most of the investors’ desired information is still not covered by regulations leaving the

decision to the company whether to address it or not. This trend of increased disclosure

driven both by companies and external bodies paired with a lower cost of obtaining

information driven by technological progress results in the current disclosure

environment where investors have access to longer and more extensive annual reports

(KPMG, 2016; Deloitte, 2017; ACCA, 2012), transcribed earnings calls or financial press

releases. However, when it comes to disclosure, the more is not always the merrier: as

some studies highlight, additional information can lead to information overload (Schick

et al., 1990). Nonetheless, the general consensus among users of financial reports is not

that they want less but rather more meaningful information to be disclosed and

potentially better presented (EY, 2014; KPMG, 2011; Paredes, 2013).

2.3.2 Disclosure & LEGCs

The existing theory on corporate disclosure, both mandatory and voluntary, is very well

established with extensive studies on its relevance and economic impact. Yet, it is

important to note that most of the studies are conducted in the period 1990 to 2005 and

their samples include listed and mature companies, mostly in conservative industries

such as manufacturing (e.g. Botosan, 1997). Similar to what is aimed for with the LEGCs

samples in this thesis, Gray & Skogsvik (2004) focused on voluntary disclosure done by

pharmaceutical companies during the period 1984-1998, since the dynamics of their

business model, as well as the uncertainty about their future performance, made them

an interesting case study to look into at that point in time.

The current business environment is shifting significantly and an emerging term in

management literature to characterise it is VUCA – volatile, uncertain, complex and

ambiguous (Bennett & Lemoine, 2014), in turn leading to a shift in demand and business

models serving the needs of this business environment. Consequently, the need to verify

the validity and the applicability of previous literature results in this new environment

and on new firms appears to be high.

of 95

First of all, it is important to analyse the credibility of the information. Multiple studies

show that reliability and believability is crucial for analysts and investors to react to the

information provided to them (Sobel, 1985; Jennings, 1987). While financial reporting is

subject to auditing, which increases the credibility of the information (Graham et al.,

2005), voluntary disclosure is not. Companies need to find alternative ways for users of

reports to perceive the information as credible. 86.3% of CFOs consider meeting

benchmarks as a strong tool to build trust around disclosure, and 76.8% believe that

disclosing news in a timely manner also serves to build credibility on the market (Graham

et al., 2005). However, both of these credibility mechanisms are based on past

performance. LEGCs by definition have typically been on the market for a shorter period

of time and are in a less stable phase of their business life-cycle, meaning that analysts

and investors cannot rely to the same extent on their track record. This could therefore

have consequences on the credibility of information disclosed by this type of firms.

Furthermore, the existing studies on credibility of disclosure focus on earnings forecasts

or other financial metrics and little research is done on operating, strategic or business

model disclosure. One of them is a study done by Gu & Li (2007) on high-tech firms

during the ’90s suggesting that insider stock purchase reinforces credibility of voluntary

strategic disclosures.

Second, it is argued that for firms with high-growth opportunities, the necessity to reduce

information asymmetry might be more significant than for firms with low-growth

opportunities (Core, 2001). Therefore, although LEGCs could potentially be subject to

reduced mandatory disclosure requirements (e.g. SEC with EGCs or IASB with SMEs -

see Appendix A), these firms are the ones that would potentially benefit the most from

engaging in voluntary disclosure practices. Moreover, as LEGCs are younger firms which

operate in the current market environment, which is inherently pushing for more relevant

information to be disclosed as discussed in the previous section, it might be expected

that these firms are disclosing more extensive information and engage in voluntary

disclosure practices.

In conclusion, research on disclosure demonstrates that financial reporting plays a

decisive role in reducing information asymmetry when it comes to investors in the capital

market. The review about the value relevance of voluntary disclosure clearly shows that

of 95

firms’ disclosure strategies affect the market price of a firm, hence further strengthening

the idea that companies can have positive economic gains from engaging in voluntary

disclosure practices. Moreover, current trends in financial reporting show that, although

the amount of disclosure is being criticised, there is a high demand for companies to

present more comprehensive and value-relevant information in their financial reports.

Finally, with the current market environment shifting, the applicability of the previous

studies’ results to LEGCs is questioned since the underlying assumptions are being

challenged. While the outlined literature comes primarily from a capital market and

finance perspective, potentially because of data availability issues, existing literature

seems to lack the fundamental accounting perspective, which, however, constitutes the

basis for a deduced valuation model. Reverse engineering stock market valuations within

an established valuation model set-up may confront practitioners with a set of extremely

strong assumptions. It is hence reasonable to introduce a model set-up, which helps to

understand how voluntary disclosure contributes to explain the market value of LEGCs.

of 95

3 The Residual Income Valuation Set-up

In order to conceptually explain the problem of the valuation difference for LEGCs, and

the missing fundamental accounting perspective in existing literature, the RIV model is

introduced as anchor model in the valuation framework. Looking at it from a

fundamental accounting perspective, the RIV model is chosen for two major reasons:

1. Deduced valuation models, such as the RIV, allow for a meaningful assessment of

the relationship between the accounting numbers and the stock market price

2. Anchoring fundamental valuation models on book values, such as the book value

of equity in the RIV, tend to simplify prediction problems within the modelling

The model is detailed in section 3.1 in order to provide a general understanding of its

derivation and critical variables. Section 3.2 focuses on the application of RIV to LEGCs

and shows that a traditional application of RIV, with statistically well-behaved

forecasting approaches based on mandatory financial information cannot fully explain

the market value of these companies. Instead, it is shown that, in order to justify the

market value of LEGCs, abnormal forecasting assumptions as well as information

gathering procedures need to be undertaken.

3.1 Model Specification

Having in mind the upcoming phenomenon of LEGCs, a valuation set-up is necessary in

order to frame the analysis. The model chosen to approach the research question is the

RIV model, first mentioned in its basic structure by Preinreich (1938) and Edwards &

Bell (1961). The section builds primarily on Skogsvik (2002), since it provides a step-by-

step tutorial of the RIV methodology, as well as of its core value drivers. Deduced

valuation models in their essence combine accounting fundamentals with capital value

concepts from economics, thereby being conceptually build up in the following way:

Capital value = (Book value of capital) + (Present value of future abnormal (1)

earnings) + (Present value of goodwill/ badwill at the horizon point in time)

of 95

Therefore, putting this generic logic into the context of a valuation of owners’ equity,

the following expression can be derived:

𝑉0 = 𝐵0 + ∑ %&−1∗(*+,&∗ −-+)

(1+-+)&/0=1 +

%1 ∗(2131

−1)(1+-+)1 (2)

where:

𝑉0 = Capital value of owners’ equity, determined ex dividend and including

any new issue of share capital at time 𝑡

𝐵0 = Book value of owners’ equity, determined ex dividend and including any

new issue of share capital at time 𝑡

𝑅6,0∗ = Book return on owners’ equity, accrued in period 𝑡

𝜌6 = Required rate of return on owners’ equity (= cost of equity capital)

𝑇 = Horizon point in time

The function in (2) only holds if it is assumed that the Clean Surplus Relation of

Accounting (i.e. net income, dividends, new issue of share capital and share repurchases

explain changes in the book value of owners’ equity) holds:

𝐷0 − 𝑁0 = 𝐵0−1 + 𝐼0 − 𝐵0 = 𝐵0−1 ∗ 𝑅6,0

∗ − (𝐵0 − 𝐵0−1) (3) where:

𝐷0 = Expected total dividend paid to the shareholders of the company, where

t denotes time of payment

𝑁0 = Expected new issue of share capital to the company, where t denotes time

of payment

𝐼0 = Accounting net income, accrued in period t

Once the expected relative goodwill/ badwill of owners’ equity (=1 +1%1 +1

− 1) coincides with

(=1%1

− 1), the company enters the competitive equilibrium where the annual growth rate

of owners’ equity is constant and the expected book return on owners’ equity after the

horizon point in time 𝑇 can be determined under the following equilibrium condition:

of 95

𝑅6,/+∞∗ = 𝜌6 + (=1

%1− 1) ∗ (𝜌6 − 𝑔/+∞) (4)

where:

𝑔/+∞= Expected annual growth of owners’ equity after time 𝑡 = 𝑇

What can be noted in addition is that the expected book return on owners’ equity

𝑅6,/+∞∗ coincides with the required rate of return on owners’ equity 𝜌6 if either the

relative goodwill/ badwill of owners’ equity (=1%1

− 1) equals zero or the difference

between the required rate of return on owners’ equity 𝜌6 and the annual growth of

owners’ equity 𝑔/+∞ is insignificant (i.e. 𝜌6 − 𝑔/+∞ equals zero).

3.2 Prediction Problems & The Need for More Information

In order for the Residual Income Valuation model formulated in (2) to be of practical

relevance, an estimation of the input parameters is necessary, thereby trying to make

these estimates in a straight-forward but statistically robust way (Skogsvik, 2002). By

assuming that the Clean Surplus Relation of Accounting mentioned in (3) holds,

obtaining the book value of owners’ equity at present 𝐵0 is easily undertaken. Hence,

the following forecast issues initially addressed by Skogsvik (2002), and complemented

by a fourth, LEGC-specific, issue of an adequate required rate of return on owners’ equity

𝜌6 , remain:

Forecast Issue 1:

Expected future book return on owners’ equity 𝑅6,0∗ to the horizon point in time 𝑡 = 𝑇

Forecast Issue 2:

Expected relative goodwill/ badwill of owners’ equity at some ‘appropriately’ chosen

horizon point in time (=1%1

− 1)

Forecast Issue 3:

Expected future growth of owners’ equity to the horizon point in time 𝑡 = 𝑇

Forecast Issue 4:

Required rate of return on owners’ equity 𝜌6 to the horizon point in time 𝑡 = 𝑇 and in

the competitive equilibrium 𝑇 + ∞

of 95

3.2.1 Forecast Issue 1

The decisive problem, which arises by trying to understand the market price of LEGCs,

is that fundamental accounting numbers used within deduced valuation models may, at

first glance, be misleading. To make this point clearer, the illustrative graph in Figure 1

depicts distinct forecast scenarios, which provide an interesting angle about why LEGCs

have an incentive to voluntarily disclose information. The underlying reasoning is that

applying well-established forecasting methods (i.e. linear interpolation or mean reversion

of 𝑅6,0∗ ) described in Scenario I is not sufficient to entirely capture the market value of

LEGCs. By progressing from Scenario I to III, additional sources of equity value

generation are presented, thus demonstrating which strong set of model assumptions is

required to approximate stock market prices:

Figure 1 Conceptual illustration of different book returns on owners’ equity forecast scenarios

Before the explanation between the three different Forecast Issues is done, it is important

to highlight again that the horizon point in time 𝑇 is a dependent variable. Although

the three scenarios in Figure 1 truncate at the same point in time for illustrative reasons,

𝑇 is in practice derived by estimating the duration of the expected relative goodwill/

badwill of owners’ equity for each company individually, meaning that one should

truncate once the competitive equilibrium is reached.

Scenario I: Residual income forecasts of the valued firm are statistically well-behaved,

i.e. historic book returns on owners’ equity provide a first meaningful approximation for

-10%

0%

10%

20%

30%

40%

-3 -2 -1 0 1 2 ... ... T T+1

RO

E &

ρ(E

)

Time t

Linear Interpolation Accounting Measurement Bias >0 The 'Hump' Cost of Equity

Valuation Pointin Time (t = 0)

Truncation Pointin Time (T)

I

II

III

CompetitiveEquilibrium

of 95

𝑅6,0∗ at t=1, thereafter assuming, in the absence of other information, the simplified

approach of a linear gradual change from 𝑅6,1∗ to 𝑅6,/+1

∗ (which in this scenario is

assumed to be equal to 𝜌6 , i.e. the accounting measurement bias (=1%1

− 1) at the horizon

point in time is equal to 0).

Scenario II: Residual income forecasts of the valued firm show the same simplified

approach of a linear gradual change from 𝑅6,1∗ to 𝑅6,/+1

∗ , whereas 𝑅6,/ +1∗ is larger than

𝜌6 in steady state (i.e. the accounting measurement bias (=1%1

− 1) at the horizon point

in time is larger than 0), resulting in additional equity value due to the respective

competitive equilibrium conditions.

Scenario III: Residual income forecasts of the valued firm show an unusual pattern, or

in this illustration rather a distinct ‘hump’, either based on firm-specific management

forecasts or estimates by professional financial analysts, indicating that the company is

able to generate book returns on owners’ equity 𝑅6,0∗ exceeding the market level during

the forecast period (i.e. an internal rate of return for projects higher than the required

rate of return on the projects). The fundamentally higher residual incomes with their

underlying equity value need to be assessed carefully as the truncation point 𝑇 in time

with the entrance into a competitive equilibrium is dependent on this (in line with II,

𝑅6,/+1∗ is, in this illustration, larger than 𝜌6 in steady state adding another source of

equity value).

Embedding the issue raised in the context of LEGCs, it can definitely be stated that the

poor earnings performance quality of these firms in the past periods of time requires a

future trajectory similar to the one depicted in III, in order to justify the stock market

valuation. Thus, voluntarily disclosing, for example, earnings forecasts for the next

upcoming years may not only close information asymmetries, as mentioned before, but

is also necessary to control the stock price volatility.


Another input parameter crucial for the RIV model is the accounting measurement bias

(=1%1

− 1) (also known as the q-value). In line with what has been discussed in Forecast

of 95

Issue 1, (=0%0

− 1) is usually not equal to (=1%1

− 1). A relevant distinction to be done in

the context of a prediction is between business goodwill/ badwill and accounting

measurement bias, as both are key drivers of why the capital value of owners’ equity 𝑉0

is not equal to the book value of owners’ equity 𝐵0 at the valuation point in time. The

following indicative graph in Figure 2 illustrates the distinction:

Figure 2 Conceptual illustration of the q-value composition and behaviour over time for different types of firms

As briefly introduced in the course of Forecast Issue 1, the underlying logic within this

differentiation is that business goodwill/ badwill is expected to fade away over time until

it reaches the steady state equilibrium. In other words, current and upcoming projects

do not have an internal rate of return larger than the required rate of return anymore,

due to underlying market forces such as increasing competition or higher input prices.

Hence, having the truncation point in time 𝑇 at a ‘reasonably’ late state causes (=1%1

− 1)

to solely capture the accounting measurement bias, which then is a result of conservative

and prudent accounting principles (=1%1

− 1 > 0).

Considering potential business model characteristics of LEGCs, i.e. provision of services,

intense R&D activity, as well as a significant amount of intangible assets, which could

currently not entirely be captured by accounting standards, it can be said that an

assessment of (=1%1

− 1) before entering the competitive equilibrium asks for an even richer

set of accounting recognition principles applied (e.g. R&D capitalization vs. expensing or

0.0

0.5

1.0

1.5

2.0

2.5

0 1 2 3 4 5 ... ... T T+1

(V(t

)/B

(t))

-1

Time t

Mature Current Cost Accounting Firm Mature Historical Cost Accounting FirmListed Emerging Growth Company Implicit Accounting Measurement Bias

Valuation Pointin Time (t = 0)

Truncation Pointin Time (T)

CompetitiveEquilibrium

Business Goodwill/ BadwillFading Away Over Time

of 95

historical cost vs. current cost accounting). Voluntary disclosure can hereby serve as a

meaningful instrument to close the information gap.


The third prediction issue deals with the future growth of owners’ equity up to the

truncation point in time 𝑇 . Having in mind the Clean Surplus Relation of Accounting in

(3), a constant dividend policy of a firm is preferable to obtain reliable estimates of

growth rates. Looking at historical data about dividend pay-out ratio (D&E&

) or dividend

share ratio ( D&%&−1

) for a stable and mature firm may provide statistically robust estimates

for the future. What also becomes complicated for this type of firms is if events such as

further share issuances or stock repurchases occur, usually resulting in significant changes

in the book value of owners’ equity 𝐵0.

With regard to LEGCs, an estimation of the future growth rate of owners’ equity is,

again, more complicated to predict. This can be explained, as already mentioned before,

by the fact that LEGCs might be in an earlier stage of the company life cycle once they

become listed, thereby experiencing strong growth rates (e.g. on financials such as

revenue) and a more volatile business environment. Moreover, it could be exptected that

these companies either cannot afford or are not interested (e.g. due to a large set of

investment opportunities) in paying out dividends in the near future. The same applies

to share issuances and, to a lesser extent for LEGCs, share repurchases, since the timing

of these events is rather unclear. This leaves some major challenges to the valuation

exercise, given that one has to anticipate when dividend payments may start and how

they develop, as well as when a firm has an equity financing need and how much do they

intend to raise. Again, voluntary disclosure by a company’s management with

information about dividend policy and financing need may serve as an insightful device

to shed more light on these topics, therefore improving the forecast quality of the RIV

model.


Approaching the required rate of return on owners’ equity 𝜌6 is a topic of considerable

interest within academia, where the Capital Asset Pricing Model (Sharpe, 1964; Lintner,

1965; Mossin, 1966), the Dividend Discount Model (Gordon & Shapiro, 1956), or Fama

of 95

& French’s (1993) Three-Factor Model are the most common ways of obtaining this

value. Although the question of which of the outlined models is the most useful for

practitioners is not part of this thesis, this Forecast Issue still raises some important

concerns in the context of a LEGC valuation.

While it is already a rather strong assumption to hold the parameter 𝜌6 constant over

time (i.e. no significant changes in capital structure or firm-specific risk), especially the

longer the specific forecast period becomes, it is also worth investigating the role of

LEGC’s inherent bankruptcy risk. Referring back to two fundamental criteria of LEGCs,

which on the one hand is a negative book return on owners’ equity and on the other a

history of loss-making years, i.e. an accumulated deficit within the equity, the exposure

to a potential bankruptcy scenario of these firms is significant. This exposure does not

only pose the question about the expected life of these firms, but may also require a

fundamental increase in the required rate of return on owners’ equity in order to

compensate investors for the additional risk. A formula used by Skogsvik (2006) in order

to calibrate the equity cost of capital for bankruptcy risk is the following:

𝜌6,0∗ = -+,&+FGHIJ,&

(1−FGHIJ,&) (5)

where:

𝜌6,0∗ = Failure-adjusted required rate of return on owners’ equity

𝑝LMNO,0 = Probability of bankruptcy in period 𝑡, conditioned on survival at 𝑡 − 1

Putting this into the context of valuing a LEGC, the negative impact on the firm value

using a high calibrated cost of equity can be substantial, since 𝑝LMNO,0 cannot be expected

to be low for firms with this earnings performance. Besides this, the accumulated losses

within equity pose a risk on the going concern assumption of a firm and thus ultimately

also on the expected lifetime. With this being stated, it appears interesting to analyse

the data provided by LEGCs with regard to (non-)disclosure on cost of equity and

underlying driving forces as well as potential bankruptcy indicators.

To summarise, the four Forecast Issues underpinning the RIV model, presented in this

section, serve as a theoretical lens for the analyses which are conducted in the course of

this thesis.

of 95

4 Scope & Methodology

This section describes the methodology used to investigate the research question. First

the disclosure scope, as well as the differentiation between voluntary and non-voluntary

disclosure, is presented. Second, the data collection process is explained. Finally, a

description of the data analysis procedures and the validity exercises is provided.

4.1 Disclosure Scope

The disclosure analysed in this thesis focuses solely on voluntary disclosure seeing as the

IASB and FASB themselves recognise the necessity for users of financial statements to

rely on additional sources to gather the information necessary to make resource allocation

decisions (IASB, 2018, 1.6; SEC, 2018, OB.6).

Since the concept of voluntary disclosure is very broad (i.e. anything that is not required

by law) and given the research question, the term voluntary disclosure has been refined

in the following way: financial disclosure documents are analysed, excluding corporate

governance reports, CSR and sustainability reports, as well as remuneration and stock

compensation reports. Although studies highlight the value relevance that those

documents can have for a company (e.g. Dhaliwal et al., 2011), this type of disclosure is

excluded from the analysis in order to respect the scope of this thesis. Finally,

‘boilerplate’ statements, such as Safe Harbour or Force Majeure, are excluded from the

analysis as they are considered to not bring valuable information on the companies’ value

generation potential. In fact, both IASB and FASB agree that ‘boilerplate’ disclosure is

not useful and could also be used to obfuscate information (Lang & Stice-Lawrence,

2015). ‘Boilerplate’ is defined according to Lang & Stice-Lawrence (2015, p. 113) as

“standardi[s]ed disclosure that is so prevalent that it is unlikely to be informative”.

In order to distinguish between mandatory and voluntary information, the following

approach is taken: a thorough review of mandatory disclosure requirements under IFRS

(IASB, 2016) and U.S.-GAAP (FASB, 2018b; SEC, 2017) by analysing both the

standards themselves as well as disclosure guidelines issued by three of the ‘Big Four’

firms (EY, 2018; KPMG, 2017b; and PwC, 2017) is supplemented with the authors’

knowledge with regard to disclosure requirements. Moreover, for sections that are

of 95

required by the respective reporting standard but for which the content is kept

unspecified, for example Management Discussion and Analysis section in 10-K forms

required by the SEC, the content is considered as voluntary.

4.2 Data Collection

The data is manually collected by the authors. A variety of issued reports, in total 164

documents, by the selected companies are analysed (see Table 7 in Appendix B) and

include: latest available annual report, latest available quarterly report and related

documents such as earnings press release for that quarter, earnings call transcript, as

well as earnings presentation. In case one of the related documents is not available, if

possible, it is replaced by a similar document (e.g. earnings presentation replaced by

investor presentation). Although only in very few cases, companies sometimes voluntarily

issue additional documentations that either supports or complements their financial

reports; those documents are additionally analysed. An example is Netflix which issues

a presentation that explains how their content accounting works. The analysed reports

were downloaded from the respective company investor relation webpage and the

transcripts of earning calls, if not readily available on the company’s website, were either

manually transcribed or retrieved from seekingalpha.com and finance.yahoo.com.

In order to ensure a structured and systematic approach to the data collection process,

a codification matrix was designed, first in a tentative form, based on existing literature

and the authors’ understanding of what type of voluntary information can be relevant

in a deduced valuation model. One firm from the sample outlined in the following section

was randomly selected to perform a test iteration of data collection which subsequently

led to the adjustment of the codification matrix. The iterative process of reviewing the

matrix in order to match more appropriately the data was constant throughout the data

collection until all the data was fully collected and validated.

Keeping in mind the objective of capturing a holistic picture for valuation of LEGCs, the

matrix includes both Financial and Operating Information, which are so called ‘topics’

needed to perform a comprehensive valuation exercise (Penman, 2013; Koller et al.,

2015). Nonetheless, as mentioned in the literature review, information is useless if it is

of 95

not believed by its users, therefore a third topic, Credibility, is introduced in the matrix

to substantiate the presence of credibility-building statements.

The topic of Financial Information is divided into so called ‘sub-topics’, the first of which

is Residual Income Valuation Model (Skogsvik, 1998), in line with our conceptual

framework, to capture voluntary information that can be directly linked to the underlying

variables of the valuation model. During our test iteration, information related to the

Discounted Cash-Flow Model (Koller et al., 2015) was observed and therefore added as

a new sub-topic. Moreover, seeing as both models have some common variables, the third

sub-topic that has a direct valuation impact is Overlapping Model Inputs. To complete

the financial information section, a last sub-topic was added to capture Additional

Financial Information that can indirectly support the inputs necessary to complete a

valuation exercise.

The topic of Operating Information was similarly split into three sub-topics, with the

first containing company-specific Business Model information, constructed based on

Morris et al. (2005), and the remaining two containing industry-specific information

about the Competitive Landscape based on Porter’s Five Forces Model (1979) and the

Market Environment.

Finally, the topic of Credibility is split into two sub-topics: General Reputation, inspired

by Hoffmann & Fieseler (2012) and Fact Check based on Graham et al. (2005). The split

was designed to accommodate the different nature of the information, seeing as the nodes

in General Reputation can be analysed using mixed methodology as described below,

whereas Fact Check can only be analysed using qualitative analysis.

The eight sub-topics (not counting in the Fact Check sub-topic which is analysed

separately) are further divided into specific nodes; the complete matrix with definition

and examples for each of the 38 specific nodes can be found in Appendix C. The data is

collected and codified using NVivo 11 provided by QSR International, a qualitative data

analysis computer software in which the codification matrix was replicated.

of 95

4.3 Approach to Analysis

Given the concerns raised when it comes to using self-constructed proxies, scores or

measures in disclosure research, this study instead relies on an inductive mixed method

approach as it is deemed the most appropriate research method to fully investigate the

research question. Moreover, the main argument in favour of mixed method is that it

yields greater insights than monomethod studies (O’Cathain et al., 2007) regardless of

the purpose, since it allows to combine the strengths and mitigate the weaknesses of the

two traditional, qualitative and quantitative, methods (Johnson & Onwuegbuzie, 2004).

The anchor paper used for the quantitative analysis is Beattie et al. (2004), which

provides a foundation for the structure of the analyses. The anchor paper used for the

qualitative analysis is Gioia et al. (2013), which provides a rigorous methodological

approach for inductive qualitative studies. The methodologies are presented separately

for clarity purposes; however, the analyses were conducted in an iterative manner.

4.3.1 Quantitative Analysis

The first quantitative analysis can be defined as Thematic Content Analysis, meaning

that the objective is to codify text into categories (Beattie et al., 2004), or ‘nodes’ in this

thesis, and report the number of recording units identified per sub-topic, as presented

later in Table 2. This in turn allows to perform frequency analyses and determine which

sub-topics have the most recorded themes. In line with the papers reviewed by Jones &

Shoemaker (1994), this thesis also uses ‘themes’ as a recording unit measure, which gives

the highest degree of freedom to judge content, since it allows codifying anything from a

single word to a single document section as one unit. These units are referred to as

‘recorded themes’ in the remainder of this thesis.

Thematic Content Analysis can only provide information about the topics, their presence

and frequency. However, it is not sufficient to understand the quality of the disclosure

and it is definitely not sufficient to provide an answer to the research question. Therefore,

an additional level of analysis is introduced, Attribute Analysis, as suggested by Beattie

et al. (2004) which introduces two attribute classifications to the codified units: (1) Time

Attribute (distinguishing between Historical, Forward-looking or Non-specific) and

of 95

(2) Depiction Attribute (distinguishing between Qualitative, Quantitative, Both or No

Data). Moreover, the two attributes were assigned in Combinations, meaning that,

instead of listing the attributes individually, they were paired in a matrix system which

allows for a deeper understanding of the data. Figure 3 shows the complete classification

matrix for the sample firm AO World plc. Finally, the results obtained from the two

analyses on the main sample are contrasted with the control sample in order to identify

general patterns, differences and commonalities.

Figure 3 Example of a filled-out codification matrix using UK-based AO World plc

Node Historical Forward-looking Non-specific Theme CountCost of Debt No Data No Data No Data 0Leverage Ratio Qualitative No Data No Data 1Free Cash Flow No Data No Data No Data 0Underlying Free Cash Flow Line Items No Data Qualitative No Data 6Terminal Value Growth Rate No Data No Data No Data 0WACC No Data No Data Quantitative 1Cost of Equity No Data No Data No Data 0Truncation Point in Time No Data No Data No Data 0Return on Equity No Data No Data No Data 0Underlying Return on Equity Items Quantitative No Data No Data 3Growth Rate Equity Book Value Qualitative No Data No Data 1Accounting Measurement Bias No Data No Data No Data 0Supplemental Revenue Information Qualitative Qualitative No Data 3Supplemental Expense Information Qualitative Qualitative No Data 14Alternative Accounting Metrics Both Qualitative No Data 19Business Model Specific KPIs Qualitative No Data No Data 7

Strategy Execution & Consistency Qualitative Qualitative No Data 18Growth Strategy No Data Qualitative Qualitative 12Value Proposition No Data No Data Qualitative 5Mission, Vision & Values Qualitative No Data Qualitative 9M&A Strategy No Data No Data No Data 0Customer Information Both Qualitative No Data 8Employee Information Qualitative Qualitative No Data 6R&D and Project Activity No Data No Data No Data 0Bargaining Power of Suppliers Qualitative No Data Qualitative 2Bargaining Power of Customers No Data No Data No Data 0Threat of Substitutes No Data No Data No Data 0Threat of New Entrants No Data No Data No Data 0Industry Rivalry No Data No Data No Data 0Market Condition Both Both No Data 14Market Share Qualitative No Data No Data 4Innovation Trends Qualitative Qualitative No Data 1Regulatory Framework No Data No Data No Data 0

Brand Strength Both Qualitative No Data 8Public Reputation Qualitative No Data No Data 3

Sub-Total 145

Degree of Commitment Directional 11External Source Reliance Yes 2Historical Forecast Accuracy Partially Met 2

Total 160

Mar

ket

Env

iron

-m

ent

General Reputation

Fact Check

Disco

unte

d C

ash-

Flow

Mod

el

Overlapping Model Inputs

Res

idua

l In

com

e V

alua

tion

M

odel

Add

itio

nal

Fina

ncia

l D

ata

Com

petitive

La

ndsc

ape

Bus

ines

s M

odel

of 95

4.3.2 Qualitative Analysis

The qualitative analysis investigates the content codified within each specific node to

identify similarities and differences in the voluntary disclosures and their attributes

(Beattie et al., 2004). The methodology used is the one suggested by Gioia et al. (2013),

which allows the introduction of a systematic approach to inductive studies; a conceptual

illustration of the method is provided in Figure 4:

Figure 4 Conceptual illustration of the Gioia Methodology (Gioia et al., 2013)

A First Order Concept analysis is performed to explore the data based on the terminology

used by the respective companies which results in 73 First Order Concepts for the main

sample. Subsequently, a Second Order Theme analysis is completed to identify

relationships and wider trends progressing towards more theoretical Themes in the main

sample, which span across the classification matrix. This resulted in 19 Second Order

Themes. Finally, the identified concepts are refined in seven Aggregate Dimensions which

are based on theoretical notions in the field of corporate finance and which are linked

back to the Forecast Issues presented in section 3.2 in order to discuss the role that the

information plays in deduced valuation models. This process was repeated for the control

group.

1st Order Concepts 2nd Order Themes AggregateDimensions

Aggregated in Dimension 1

…

Summarized inTheme 1

§ 1st order concept X observed across sample§ 1st order concept Y observed across sample§ 1st order concept Z observed across sample

§ …§ …§ …

of 95

4.4 Data Legitimation

In order to evaluate the validity (or legitimacy, which is the term suggested by

Onwuegbuzie & Johnson (2006) for mixed methods) of the data collection and

classification, a manual validation exercise was conducted by the authors. As suggested

by Athanasakou et al. (2018), manual validation was chosen since it is an adequate test

to verify classification accuracy and reproducibility. The manual validation process was

conducted in the following way: the two authors divided the firms to be analysed,

however, initially a company was randomly selected from the sample and codified

simultaneously by the two authors. The output was then compared and revisited when

necessary in order to develop a common understanding on how to proceed with the

classification of data for the remaining firms (Gioia et al., 2013) and strengthen intercoder

reliability by ensuring different coders produce the same, or very similar, content (Beattie

et al., 2004). After the independent data collection and classification for the complete

sample was terminated, the authors swapped firms and assessed their respective outputs;

this included both a review of the nature of the data (voluntary vs. mandatory and

‘boilerplate’ vs. ‘non-boilerplate’), their attributes, as well as their classification into the

underlying nodes. Finally, the authors consolidated the data and performed a final check

of the classified output together.

of 95

5 Sample Collection

With regard to the intent and the scope of this thesis, a purposive sampling method was

used to obtain a well-suited group of firms. Contrary to random probability sampling

techniques, which represent the guiding method within quantitative research designs,

purposive sampling method is a strategic way to identify sample firms that are relevant

to the addressed research question and is highly useful to examine specific characteristics

of a population of interest. While this sampling approach does not allow for a

generalisation from sample to population, it is considered an appropriate technique for

mixed methods research designs which try to a understand a phenomenon (Bryman &

Bell, 2011).

5.1 Collection of Main Sample

Considering the scope of this thesis along with the respective disclosure documents that

are analysed for each firm, a total sample size of 25 LEGCs is defined, given that the

authors regard this as a sufficient number to determine patterns and characteristics, thus

enabling them to draw relevant conclusions.

With the chosen sampling method in mind, it is necessary to establish a certain set of

criteria so that it becomes clear which companies do qualify for inclusion in the main

sample and which do not. Referring back to the definition of LEGCs, the following

metrics and criteria are set-up in order to obtain the main sample:

§ Stock market listing with free-floating and liquid shares;

§ Traded for more than one and less than ten successive years;

§ Operations in an emerging industry as defined by PwC (2012), belonging to the

technology sector;

§ Three-year revenue CAGR of at least 10%;

§ Negative book return on owners’ equity as of the latest audited financial

statements; and

§ History of loss-making years, i.e. accumulated deficit within the equity as of the

latest audited financial statements.

of 95

With these predefined criteria in mind, the sample collection was approached using the

Thomson Reuters database Eikon starting off with a data request which provided a list

with all IPOs between the beginning of 2008 and the end of 2017 at all major stock

markets in the USA (NASDAQ and New York Stock Exchange) as well as Europe

(Frankfurt Stock Exchange, London Stock Exchange, Euronext, NASDAQ Nordic). The

results were then filtered according to the aforementioned criteria and 25, among all

appropriate companies, were selected. This selection was based on size and relevance of

the firm for the purpose of this thesis.

Another criterion, which is not part of the LEGC definition but considered to be

meaningful in the sample collection, is the reporting standards under which the

companies present their financial reports. Hence, the main sample was split up into 13

LEGCs reporting under U.S.-GAAP and 12 LEGCs reporting under IFRS, since these

are the predominant reporting standards at the chosen stock exchanges. Although the

two reporting standards have some differences, their level of convergence is considerably

high, making the mandatory information reported by companies comparable (PwC,

2018).

5.2 Collection of Control Sample

In order to benchmark the findings about LEGCs’ voluntary disclosure in the main

sample, a control group is set-up consisting of sample firms that represent the ‘optimal-

but-profitable twin’ to the LEGCs defined in the main sample. This means that the

control firms also are listed, have been trading with liquid and free-floating shares for

more than one year, operate in similar industries and have a three-year revenue CAGR

of at least 10%. Since the objective is to control for profitability, the following two criteria

replace the negative book return on equity, as well as the accumulated deficit:

§ Positive book return on owners’ equity as of the latest audited financial

statements; and

§ History of profit-making years, i.e. retained earnings within the equity as of the

latest audited financial statements.

of 95

Given the time constraints of this thesis, a control group of 10 companies is perceived to

be large enough to control for differences. Comparable to the notion applied in the main

sample, the control group is equally split up into firms reporting under IFRS and U.S.-

GAAP. The final control sample constitutes of 5 U.S.-GAAP and 5 IFRS companies.

Table 1 - Sample Overview Busin

ess D

escr

iptio

n

Repo

rtin

g St

anda

rd

Coun

try

of In

corp

orat

ion

IPO

Year

Prim

ary

Stoc

k Ex

chan

ge

Late

st R

epor

ting

Date

Mar

ket C

apta

lisat

ion[in

mUS

D]

Main Sample1 AO World plc E-Commerce IFRS UK 2014 LSE 31.03.2018 65.32 Atlassian Corporation Plc SaaS IFRS Australia 2015 NASDAQ 30.06.2018 14,711.93 Boozt AB E-Commerce IFRS Sweden 2017 NASDAQ Nordic 30.06.2018 457.14 Box, Inc. SaaS U.S.-GAAP USA 2015 NYSE 31.07.2018 3,392.95 Delivery Hero SE Food Tech IFRS Germany 2017 FSE 30.06.2018 9,871.26 HelloFresh SE Food Tech IFRS Germany 2017 FSE 30.06.2018 2,460.07 LendingClub Corporation FinTech U.S.-GAAP USA 2014 NYSE 30.06.2018 1,603.98 LINE Corporation Social Media IFRS Japan 2016 NYSE 30.06.2018 9,778.99 MyBucks S.A. FinTech IFRS Luxembourg 2016 FSE 31.12.2017 113.7

10 Okta, Inc. SaaS U.S.-GAAP USA 2017 NASDAQ 31.07.2018 5,411.611 OnDeck Capital, Inc. FinTech U.S.-GAAP USA 2014 NYSE 30.06.2018 522.512 ServiceNow, Inc. SaaS U.S.-GAAP USA 2012 NYSE 30.06.2018 30,682.413 Shop Apotheke Europe N.V. E-Commerce IFRS Netherlands 2016 FSE 30.06.2018 623.114 Shopify, Inc. SaaS U.S.-GAAP Canada 2015 NYSE 30.06.2018 15,518.315 Snap Inc. Social Media U.S.-GAAP USA 2017 NYSE 30.06.2018 12,550.716 Splunk, Inc. SaaS U.S.-GAAP USA 2012 NASDAQ 31.07.2018 14,089.417 Square, Inc. FinTech U.S.-GAAP USA 2015 NASDAQ 30.06.2018 25,084.518 Tableau Software, Inc. SaaS U.S.-GAAP USA 2013 NYSE 30.06.2018 8,107.419 Takeaway.com N.V. Food Tech IFRS Netherlands 2016 Euronext 30.06.2018 2,877.120 Talend S.A. SaaS IFRS USA 2016 NASDAQ 30.06.2018 1,846.021 Twilio Inc. CPaaS U.S.-GAAP USA 2016 NYSE 30.06.2018 5,453.022 Twitter, Inc. Social Media U.S.-GAAP USA 2013 NYSE 30.06.2018 33,056.323 windeln.de SE E-Commerce IFRS Germany 2015 FSE 30.06.2018 46.424 Workday, Inc. SaaS U.S.-GAAP USA 2012 NASDAQ 31.07.2018 26,912.325 Xero Limited SaaS IFRS New Zealand 2012 ASX 31.03.2018 3,753.5

Control Sample1 Catena Media p.l.c Ad Tech IFRS Malta 2016 NASDAQ Nordic 30.06.2018 5,771.72 Grubhub Inc. Food Tech U.S.-GAAP USA 2014 NYSE 30.06.2018 9,477.33 Micro Focus International plc SaaS IFRS UK 2005 LSE 30.04.2018 913,305.24 Netflix, Inc. VoD U.S.-GAAP USA 2002 NASDAQ 30.06.2018 170,451.15 Paypal Holdings, Inc. FinTech U.S.-GAAP USA 2015 NASDAQ 30.06.2018 98,591.76 Rovio Entertainment Oyj GaaS IFRS Finland 2017 NASDAQ Nordic 30.06.2018 499.47 salesforce.com, inc. SaaS U.S.-GAAP USA 2004 NYSE 31.07.2018 103,781.48 Scout24 AG Marketplace IFRS Germany 2015 FSE 30.06.2018 5,695.09 Veeva Systems Inc. SaaS U.S.-GAAP USA 2013 NYSE 31.07.2018 9,214.0

10 Zalando SE E-Commerce IFRS Germany 2014 FSE 30.06.2018 13,744.4Table 1 depicts the firms which constitute the main sample and control sample along with additional company information such as Business Description , Reporting Standard , Country of Incorporation , IPO Year , Primary Stock Exchange , Latest Reporting Date and Market Capitalisation [in mUSD] . Market capitalisation is as of the latest reporting date of the respective financial disclosure documents used in the analysis of this thesis. The control sample controls for profitability, i.e. the companies have a positive book return on owners' equity as well as retained earnings in contrast to the companies constituting the main sample .

of 95

6 Data Description & Analysis

6.1 Quantitative Analysis

This sub-section represents the first step in describing and analysing the voluntary

disclosure of LEGCs by comparing the distribution of recorded themes and the attribute

combinations, i.e. the combination of Time Attributes (Historical, Forward-looking or

Non-specific) with Depiction Attributes (Qualitative, Quantitative, Both or No Data),

across sub-topics and nodes. As already outlined in sub-section 4.3.1, the voluntary

disclosure practice of the 25 LEGCs constituting the main sample is controlled for by

looking into the disclosure practice of 10 profitable but otherwise comparable ‘twin

LEGCs’, which represent the control sample. Given the differences in sample size, the

output is reported in percentages to allow comparability.

6.1.1 Recorded Theme Distributions & Analysis

Starting off with Table 2, the total number of recorded themes from the content analysis

is presented, thereby also showing the distribution across the eight sub-topics outlined

in section 4.2. The total number of recorded themes in the main sample is 3,597, whereas

there are 1,574 in the control sample. Given the difference in size between main and

control sample (i.e. with the control sample representing 40% of the main sample), no

major difference in the overall quantity can be determined.

However, moving on to the more granular sub-topic level, it can be observed for both

samples that the vast majority of recorded themes are in the Additional Financial Data

(37.1% and 36.5% respectively) and the Business Model sub-topic (31.7% and 28.8%

respectively). Market Environment (11.1% and 11.8% respectively) and Discounted Cash-

Flow (6.3% and 10.0% respectively) represent the third and fourth largest sub-topic

although there is already a notable difference with respect to the ones mentioned before.

Since Total Number of Recorded Themes is a flawed measure, meaning that it does not

capture the difference in the total number of underlying nodes within each sub-topic,

Mean Recorded Themes per Underlying Node is introduced to adjust for this effect.

Irrespective of the underlying number of nodes, Additional Financial Data (333.8 and

143.5 respectively) and Business Model (142.4 and 56.8 respectively) remain the sub-

topics with the highest number of recorded themes per node.

of 95

Tab

le 2

- D

escr

iptiv

e St

atist

ics

of R

ecor

ded

The

mes

M

ain

Sam

ple

Tot

alC

ount

Disc

ount

ed

Cash

-Flo

wO

verla

ppin

g M

odel

Inpu

tsR

esid

ual I

ncom

e V

alua

tion

Add

ition

al

Fina

ncia

l Dat

aBu

sines

sM

odel

Com

petit

ive

Land

scap

eM

arke

t En

viro

nmen

tG

ener

al

Rep

utat

ion

Tota

l Num

ber

of R

ecor

ded

Them

es3,

597

227

417

91,

335

1,13

920

640

010

7%

of T

otal

6.3%

0.1%

5.0%

37.1

%31

.7%

5.7%

11.1

%3.

0%N

umbe

r of

Rec

orde

d Th

emes

62

44

85

42

Mea

n R

ecor

ded

Them

es p

er U

nder

lyin

g N

ode

37.8

2.0

44.8

333.

814

2.4

41.2

100.

053

.5

Mea

n R

ecor

ded

Them

es p

er F

irm14

3.9

9.1

0.2

7.2

53.4

45.6

8.2

16.0

4.3

Med

ian

Rec

orde

d Th

emes

per

Firm

145.

09.

00.

05.

053

.049

.08.

015

.04.

0M

axim

um R

ecor

ded

Them

es in

a F

irm21

924

229

9166

2341

17M

inim

um R

ecor

ded

Them

es in

a F

irm86

00

021

160

30

Num

ber

of S

ampl

e Fi

rms

25

C

ontr

ol S

ampl

eT

otal

Cou

ntD

iscou

nted

Ca

sh-F

low

Ove

rlapp

ing

Mod

el In

puts

Res

idua

l Inc

ome

Val

uatio

nA

dditi

onal

Fi

nanc

ial D

ata

Busin

ess

Mod

elCo

mpe

titiv

e La

ndsc

ape

Mar

ket

Envi

ronm

ent

Gen

eral

R

eput

atio

nTo

tal N

umbe

r of

Rec

orde

d Th

emes

1,57

415

70

106

574

454

6218

536

% o

f Tot

al10

.0%

-6.

7%36

.5%

28.8

%3.

9%11

.8%

2.3%

Num

ber

of R

ecor

ded

Them

es6

24

48

54

2M

ean

Rec

orde

d Th

emes

per

Und

erly

ing

Nod

e26

.2-

26.5

143.

556

.812

.446

.318

.0

Mea

n R

ecor

ded

Them

es p

er F

irm15

7.4

15.7

0.0

10.6

57.4

45.4

6.2

18.5

3.6

Med

ian

Rec

orde

d Th

emes

per

Firm

160.

012

.50.

010

.562

.548

.54.

016

.04.

0M

axim

um R

ecor

ded

Them

es in

a F

irm21

040

023

8366

1838

7M

inim

um R

ecor

ded

Them

es in

a F

irm98

00

029

280

80

Num

ber

of S

ampl

e Fi

rms

10T

able

2 p

rovi

des d

escr

iptiv

e sta

tistic

s of t

he to

tal r

ecor

ded

them

es a

cros

s 25

firm

s whi

ch a

re p

art o

f the

mai

n sa

mpl

e and

10

firm

s whi

ch a

re p

art o

f the

cont

rol s

ampl

e. A

reco

rded

them

e can

rang

e fro

m a

sing

le wo

rd to

a si

ngle

docu

men

t sec

tion

as a

pot

entia

l re

cord

ing

unit.

The

codi

ficat

ion

mat

rix a

pplie

d is

sepa

rate

d in

to th

ree t

opics

: (1)

Fin

anci

al I

nfor

mat

ion

, (2)

Ope

ratin

g In

form

atio

n a

nd (3

) Cre

dibi

lity

with

Disc

ount

ed C

ash-

Flow

Mod

el, O

verla

ppin

g M

odel

Inp

uts,

Res

idua

l Inc

ome

Valu

atio

n M

odel

and

Ad

ditio

nal F

inan

cial

Dat

a as

sub-

topi

cs b

elong

ing

to (1

), Bu

sines

s M

odel

, Com

petit

ive

Land

scap

e an

d M

arke

t Env

ironm

ent

as su

b-to

pics

belo

ngin

g to

(2) a

nd G

ener

al R

eput

atio

n a

s sub

-topi

c belo

ngin

g to

(3).

Mea

n Re

cord

ed T

hem

es p

er U

nder

lyin

g No

de is

intr

oduc

ed a

s a m

easu

re o

f com

para

bilit

y du

e to

diffe

renc

es in

the n

umbe

r of n

odes

und

erly

ing

a su

b-to

pic.

Mea

n Re

cord

ed T

hem

es, M

edia

n Re

cord

ed T

hem

es, M

axim

um R

ecor

ded

Them

es a

nd M

inim

um R

ecor

ded

Them

es p

rovi

de fu

rthe

r dat

a po

ints

on

a fi

rm le

vel.

of 95

The considerable commonality between the main and control sample highlights that,

regardless of their profitability, LEGCs focus their voluntary disclosure on explaining to

users how their business works and what their overall potential is, both in terms of

financial results, as well as strategy. Interpreting this finding might lead to the conclusion

that voluntary disclosure is undertaken due to the novelty and the complexity of these

business models and offerings paired with the lack of proven track record on the stock

market.

Although it is apparent that the recorded themes distribution between the two samples

is very similar across the eight sub-topics, there are three differences that are important

to note: firstly, the Financial Information topic constitutes the majority of recorded

themes for the control sample, 53.2%, contrary to the main sample for which it represents

solely 48.5%. Secondly, the control group discloses more information which are directly

linked to a valuation model, i.e. the first three sub-topics, than the main group (16.7%

and 11.4% respectively). Lastly, the companies in the control sample disclose less

information belonging to the Competitive Landscape sub-topic than the companies in the

main sample do (3.9% and 5.7% respectively). This could suggest that profitable LEGCs,

which have already reached a more stable situation from a financial standpoint, can

provide more solid information about their value generation potential and have a lesser

need to discuss their market positioning.

The sub-topic of Overlapping Model Inputs which comprises the Cost of Equity as well

as the Truncation Point in Time node can be summarised as an area where almost no

voluntary information is presented within both samples. For both nodes it seems to be

fair to expect an argumentation in line with Wagenhofer (1990), given that a company

neither wants to disclose to the market at what point in time they expect to enter into

a competitive equilibrium, i.e. have reached the point in time in which they are growing

with the market, nor do they want to disclose what their own assessment of the required

rate of return on owners’ equity is. This already hints to the necessity of a broader

discussion about what considerations may impact managerial behaviour in voluntary

disclosure strategies (Verrecchia, 2001).

of 95

Since this level of aggregation lacks some analytical depth, the two sub-topics with the

highest amount of recorded themes per underlying node, i.e. Additional Financial Data

(AFD) and Business Model (BM), are further investigated looking into the percentage

distribution of the recorded themes across the underlying nodes. As can be seen in Table

3, the most frequently discussed issues in the main sample are respectively Business

Model Specific KPIs (30.9%) and Alternative Accounting Metrics (25.6%) as well as

Strategy Execution & Consistency (26.3%) and Value Proposition (19.0%). The

distribution is very similar for the control sample with two differences worth mentioning:

firstly, companies in the main sample disclose more Supplemental Expense Information

than the control sample (23.0% and 14.1% respectively), which in turn is offset by more

recorded themes for the control group in Alternative Accounting Metrics (25.6% and

33.4% respectively). Secondly, the difference in recorded themes in M&A Strategy can be

highlighted given that the main sample discloses less in this node than the control sample

(2.9% and 6.6% respectively).

Considering the relatively higher proportion of Supplemental Expense Information

recorded themes in the main sample, it could be inferred that LEGCs in the main sample

see the need to voluntarily inform more about the underlying reasons why their business

Table 3 - Number of Recorded Themes in Specific NodesMain

Sample% of Sub-

Topic TotalControl Sample

% of Sub-Topic Total

Additional Financial Data 1,335 574Supplemental Revenue Information 274 20.5% 125 21.8%Supplemental Expense Information 307 23.0% 81 14.1%Alternative Accounting Metrics 342 25.6% 192 33.4%Business Model Specific KPIs 412 30.9% 176 30.7%

Business Model 1,139 454Strategy Execution & Consistency 300 26.3% 136 30.0%Growth Strategy 158 13.9% 68 15.0%Value Proposition 216 19.0% 91 20.0%Mission, Vision & Values 121 10.6% 27 5.9%M&A Strategy 33 2.9% 30 6.6%Customer Information 135 11.9% 47 10.4%Employee Information 96 8.4% 24 5.3%R&D and Project Activity 80 7.0% 31 6.8%Table 3 provides the distribution of the number of recorded themes across specific nodes. The nodes depicted belong to the two sub-topics with the highest amount of recorded themes, i.e. Additional Financial Data and Business Model .

of 95

is unprofitable compared to the ones in the control group, thereby potentially being

aware of the negative stock price impact of missing earnings benchmark, which is in line

with the findings of Skinner & Sloan (2002). Furthermore, putting the difference in the

M&A Strategy node into context, it could be claimed that M&A scenarios of any type

are rather a topic of interest for a financially stable firm than for a non-profitable firm,

hence justifying the difference observed. What is surprising, though, is that the majority

of the financial information provided falls, for both samples, into the Business Model

Specific KPIs and Alternative Accounting Metrics nodes, which do not contain the type

of information one could directly use for deduced valuation models such as the RIV.

Although the recorded theme distribution sheds some light on the data collected, the

conclusions that can be generated appear insufficient to draw concrete findings, thus

requiring another, more detailed level of analysis.

6.1.2 Attribute Combinations & Sub-topic Level Analysis

Looking into Table 4, the concept of attribute combinations is introduced, providing the

starting point for a more granular quantitative analysis of the collected data. Within the

already established notion of attributes, a distinction can be made between Time

Attributes, i.e. which point in time the recorded theme is referring to (Historical,

Forward-looking or Non-Specific), and Depiction Attributes, i.e. how the recorded theme

is presented (Qualitative, Quantitative, Both or No Data). The data presented in Table

4 are aggregated on a sub-topic level and show the distribution of attribute combinations

across available data, i.e. excluding classifications in No Data.

Focusing on the sub-topics of Discounted Cash-Flow Model (DCF) and Residual Income

Valuation Model (RIV), the data collected in the main sample are described first. The

largest portion can be found in the attribute combination Qualitative / Forward-looking

with 32.6% of available data in DCF and 32.9% in RIV. The second most common

intersection is Quantitative / Historical for DCF with 17.4% and Qualitative / Historical

with 23.7% for RIV. Moving to the control sample, a notable difference can be seen in

the DCF sub-topic with regard to the quantitative depiction attribute, where in total

61.5% of available data can be found (33.3% in Quantitative / Historical and 28.2% in

Quantitative / Forward-looking). The strong distribution to the quantitative depiction

attribute can also be found for RIV in the Quantitative / Historical intersection with

of 95

26.5% but clearly less in Quantitative / Forward-looking (8.8%). Given the fact that the

control sample is controlling for profitability, it appears reasonable to observe more

numerical data as these firms have already reached a more stable and profit-generating

stage in which it may become easier to provide specific, i.e. quantitative, and in the case

of DCF also forward-looking, statements. What is noteworthy here is that the control

sample has a much higher distribution in the forward-looking time attribute for DCF

than for RIV, ultimately because Free Cash Flow data (and underlying items) are more

disclosed than ROE data (and underlying items).

Referring back to the sub-topics chosen for Table 3 specifically, Additional Financial

Data (AFD) and Business Model (BM) and looking at Table 4, it can be said that these

are not only the sub-topics with the most recorded themes, but also with the relatively

highest amount of available data, i.e. specific attribute combinations (AFD: 57.0% and

56.7% respectively, BM: 48.7% and 48.3%). Comparing the main with the control sample

for the AFD sub-topic, it emerges that most attribute combinations are classified into

the depiction attribute Both (50.8% and 57.3% respectively) with the majority in the

attribute combination Both / Historical (32.7% and 33.8% respectively). However, to

draw further conclusion about this sub-topic, a more granular analysis is necessary. With

no major differences between main and control sample, the vast majority of available

data being Qualitative (in total 90.1% and 87.9% respectively) and a fairly equal

distribution across the time attributes, the BM sub-topic does not appear to be very

insightful at first glance. This is especially driven by a problem which the three remaining

sub-topics, i.e. Competitive Landscape, Market Environment and General Reputation,

also encounter. Since all of these sub-topics with their underlying nodes belong to the

operating information topic, it is not unexpected that a substantial part of the codified

themes belongs to the Qualitative depiction attribute because of the inherently

qualitative nature of these data. One interesting finding is that for both the main and

the control group Market Environment is one of the nodes with the highest amount of

available data (40.3% and 43.3% respectively).

Having said that, the subsequent section adds one final level of analysis which means

that the AFD and the BM sub-topic are investigated on a node level, as they are the

sub-topics with the relatively highest amount of available data.

of 95

Tab

le 4

- D

istrib

utio

n of

Att

ribut

e C

ombi

natio

ns w

ithin

Sub

-top

ics

M

ain

Sam

ple

Con

trol

Sam

ple

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Dis

coun

ted

Cas

h-Fl

ow M

odel

450

180

No

Dat

a36

414

1A

vaila

ble

Dat

a86

39Q

ualit

ativ

e37

8.1%

32.6

%2.

3%4

2.6%

5.1%

2.6%

Qua

ntita

tive

2817

.4%

10.5

%4.

7%24

33.3

%28

.2%

-Bo

th21

12.8

%10

.5%

1.2%

1112

.8%

15.4

%-

Ove

rlap

ping

Mod

el I

nput

s15

060

No

Dat

a14

760

Ava

ilabl

e D

ata

30

Qua

litat

ive

233

.3%

-33

.3%

0-

--

Qua

ntita

tive

1-

33.3

%-

0-

--

Both

0-

--

0-

--

Res

idua

l In

com

e V

alua

tion

Mod

el30

012

0N

o D

ata

224

86A

vaila

ble

Dat

a76

34Q

ualit

ativ

e50

23.7

%32

.9%

9.2%

1717

.6%

14.7

%17

.6%

Qua

ntita

tive

2219

.7%

9.2%

-12

26.5

%8.

8%-

Both

41.

3%3.

9%-

55.

9%8.

8%-

Add

itio

nal

Fina

ncia

l D

ata

300

120

No

Dat

a12

952

Ava

ilabl

e D

ata

171

68Q

ualit

ativ

e57

11.1

%21

.6%

0.6%

137.

4%10

.3%

1.5%

Qua

ntita

tive

279.

9%5.

8%-

1610

.3%

13.2

%-

Both

8732

.7%

18.1

%-

3933

.8%

23.5

%-

Tab

le 4

see

subs

eque

nt p

age

for

deta

iled

tabl

e de

scrip

tion

.

of 95

Tab

le 4

- co

ntin

ued

M

ain

Sam

ple

Con

trol

Sam

ple

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Bus

ines

s M

odel

600

240

No

Dat

a30

812

4A

vaila

ble

Dat

a29

211

6Q

ualit

ativ

e26

324

.7%

27.4

%38

.0%

102

27.6

%26

.7%

33.6

%Q

uant

itativ

e4

0.7%

0.7%

-7

4.3%

1.7%

-Bo

th25

6.2%

2.1%

0.3%

73.

4%1.

7%0.

9%

Com

peti

tive

Lan

dsca

pe37

515

0N

o D

ata

292

128

Ava

ilabl

e D

ata

8322

Qua

litat

ive

7910

.8%

16.9

%67

.5%

2113

.6%

4.5%

77.3

%Q

uant

itativ

e0

--

-1

4.5%

--

Both

42.

4%-

2.4%

0-

--

Mar

ket

Env

iron

men

t30

012

0N

o D

ata

179

68A

vaila

ble

Dat

a12

152

Qua

litat

ive

8419

.8%

19.8

%29

.8%

3725

.0%

28.8

%17

.3%

Qua

ntita

tive

63.

3%1.

7%-

41.

9%5.

8%-

Both

3114

.0%

11.6

%-

1115

.4%

5.8%

-

Gen

eral

Rep

utat

ion

150

60N

o D

ata

100

46A

vaila

ble

Dat

a50

14Q

ualit

ativ

e43

38.0

%12

.0%

36.0

%14

64.3

%7.

1%28

.6%

Qua

ntita

tive

12.

0%-

-0

--

-Bo

th6

12.0

%-

-0

--

-T

able

4 p

rovi

des t

he d

istrib

utio

n of

att

ribut

e com

bina

tions

with

in th

e eig

ht su

b-to

pics

: Disc

ount

ed C

ash-

Flow

Mod

el, O

verla

ppin

g M

odel

Inp

ut, R

esid

ual I

ncom

e Va

luat

ion

Mod

el, A

dditi

onal

Fin

anci

al D

ata,

Bus

ines

s M

odel

, Co

mpe

titiv

e La

ndsc

ape,

Mar

ket E

niro

nmen

t an

d G

ener

al R

eput

atio

n. A

ttrib

ute c

ombi

natio

n m

eans

each

pos

sible

com

bina

tion

of th

e thr

ee ti

me a

ttrib

utes

, i.e.

Hist

oric

al, F

orwa

rd-lo

okin

g an

d No

n-sp

ecifi

c, w

ith th

e fou

r dep

ictio

n at

trib

utes

, i.e.

No

Data

, Qua

litat

ive,

Qua

ntita

tive

and

Both

(i.e.

Qua

litat

ive

and

Qua

ntita

tive)

. No

Data

illu

stra

tes i

n wh

ich su

b-to

pics

vol

unta

ry in

form

atio

n is

not d

isclo

sed

ther

eby

bein

g m

utua

lly ex

clusiv

e with

a cl

assif

icatio

n in

eith

er

Qua

litat

ive

or Q

uant

itativ

e or

Bot

h. T

his l

eads

to tw

elve d

iffer

ent c

ombi

natio

n sc

enar

ios (

thre

e tim

e att

ribut

es m

ultip

lied

with

four

dep

ictio

n at

trib

utes

), wh

ereb

y ea

ch fi

rm a

lway

s has

solel

y th

ree c

lass

ifica

tions

per

nod

e. An

exam

ple f

or

how

to re

ad th

is ta

ble i

s tha

t acr

oss 2

5 m

ain

sam

ple f

irms,

8.1%

of t

he 8

6 at

trib

ute c

lass

ifica

tions

for t

he D

iscou

nted

Cas

h-Fl

ow M

odel

sub-

topi

c (ex

cludi

ng 3

64 N

o Da

ta cl

assif

icatio

ns, i

.e. 4

50 m

inus

364

) are

Qua

litat

ive

(dep

ictio

n at

trib

ute)

/ H

istor

ical

(tim

e att

ribut

e).

of 95

6.1.3 Attribute Combinations & Node Level Analysis

The distribution of attribute combinations in Table 5 and 6 follows the same logic which

has already been established at the beginning of sub-section 6.1.2. The quantitative

analysis obtains its final and most detailed level by comparing the attribute combinations

within specifically chosen sub-topics, i.e. AFD and BM, on an underlying node level.

In Table 5, special emphasis is first laid on the Supplemental Expense Information node.

With 31.1% of the available data in Both / Historical and 37.8% in

Qualitative / Forward-looking, the main sample stands out in comparison to the control

sample where the distribution across depiction and time attributes is a lot more uniform.

Again, by controlling for profitability within the control sample, it should not come as a

surprise that companies in the main sample disclose more voluntarily, qualitatively as

well as quantitatively, about expenses to explain and justify their loss-making past

compared to companies in the control sample. Building up on this point, it also appears

reasonable that unprofitable LEGCs tend to report more qualitatively about the future

development of expenses as they want to provide the reader with explanations about the

nature and the expected development of upcoming spending by the company.

With the majority of recorded themes falling into Alternative Accounting Metrics and

Business Model Specific KPIs, the focus shifts on these two nodes. The patterns within

the first mentioned node are fairly equal between main and control sample, meaning that

the large part of attribute combinations is in Quantitative / Historical and

Both / Historical (in total 56.1% and 50.0% respectively) and most of the Forward-

looking data in Both (24.4% and 27.8% respectively). This seems to indicate that

independent of profitability, LEGCs are using a lot of these metrics in their disclosure

practice. Business Model Specific KPIs has an apparent overlap in the main and control

sample at the Both / Historical intersection (59.5% and 56.3% respectively). Next to the

unexpectedly high distribution of available data in these nodes, this is empirically

surprising and making a logical inference is not intuitive at first. Ultimately, a more

thorough analysis of the codified units is necessary to come up with an explanation for

this.

of 95

Tab

le 5

- D

istrib

utio

n of

Att

ribut

e C

ombi

natio

ns w

ithin

Add

ition

al F

inan

cial

Dat

a N

odes

M

ain

Sam

ple

Con

trol

Sam

ple

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Supp

lem

enta

l R

even

ue I

nfor

mat

ion

7530

No

Dat

a27

13A

vaila

ble

Dat

a48

17Q

ualit

ativ

e19

27.1

%10

.4%

2.1%

35.

9%5.

9%5.

9%Q

uant

itativ

e4

2.1%

6.3%

-3

-17

.6%

-Bo

th25

18.8

%33

.3%

-11

35.3

%29

.4%

-

Supp

lem

enta

l Exp

ense

Inf

orm

atio

n75

30N

o D

ata

3013

Ava

ilabl

e D

ata

4517

Qua

litat

ive

2211

.1%

37.8

%-

723

.5%

17.6

%-

Qua

ntita

tive

58.

9%2.

2%-

417

.6%

5.9%

-Bo

th18

31.1

%8.

9%-

611

.8%

23.5

%-

Alt

erna

tive

Acc

ount

ing

Met

rics

7530

No

Dat

a34

12A

vaila

ble

Dat

a41

18Q

ualit

ativ

e5

-12

.2%

-1

-5.

6%-

Qua

ntita

tive

1529

.3%

7.3%

-6

16.7

%16

.7%

-Bo

th21

26.8

%24

.4%

-11

33.3

%27

.8%

-

Bus

ines

s M

odel

Spe

cific

KP

Is75

30N

o D

ata

3814

Ava

ilabl

e D

ata

3716

Qua

litat

ive

112.

7%27

.0%

-2

-12

.5%

-Q

uant

itativ

e3

-8.

1%-

36.

3%12

.5%

-Bo

th23

59.5

%2.

7%-

1156

.3%

12.5

%-

Tab

le 5

pro

vide

s the

dist

ribut

ion

of a

ttrib

ute c

ombi

natio

ns w

ithin

the f

our n

odes

und

erly

ing

the A

dditi

onal

Fin

anci

al D

ata

sub-

topi

c: Su

pple

men

tal R

even

ue I

nfor

mat

ion

, Sup

plem

enta

l Exp

ense

Inf

orm

atio

n, A

ltern

ativ

e Ac

coun

ting

Met

rics

and

Busin

ess

Mod

el S

peci

fic K

PIs.

Att

ribut

e com

bina

tion

mea

ns ea

ch p

ossib

le co

mbi

natio

n of

the t

hree

tim

e att

ribut

es, i

.e. H

istor

ical

, For

ward

-look

ing

and

Non-

spec

ific,

with

the f

our d

epict

ion

attr

ibut

es, i

.e. N

o Da

ta,

Qua

litat

ive,

Qua

ntita

tive

and

Both

(i.e.

Qua

litat

ive

and

Qua

ntita

tive)

. No

Data

illu

stra

tes i

n wh

ich n

odes

vol

unta

ry in

form

atio

n is

not d

isclo

sed

ther

eby

bein

g m

utua

lly ex

clusiv

e with

a cl

assif

icatio

n in

eith

er Q

ualit

ativ

e or

Qua

ntita

tive

or B

oth.

Thi

s lea

ds to

twelv

e diff

eren

t com

bina

tion

scen

ario

s (th

ree t

ime a

ttrib

utes

mul

tiplie

d wi

th fo

ur d

epict

ion

attr

ibut

es),

wher

eby

each

firm

alw

ays h

as so

lely

thre

e cla

ssifi

catio

ns p

er n

ode.

An ex

ampl

e for

how

to re

ad th

is ta

ble i

s tha

t ac

ross

25

mai

n sa

mpl

e firm

s, 27

.1%

of t

he 4

8 at

trib

ute c

lass

ifica

tions

for t

he S

uppl

emen

tal R

even

ue I

nfor

mat

ion

nod

e (ex

cludi

ng 2

7 No

Dat

a cla

ssifi

catio

ns, i

.e. 7

5 m

inus

27)

are

Qua

litat

ive

(dep

ictio

n at

trib

ute)

/ H

istor

ical

(tim

e at

trib

ute)

.

of 95

Investigating Table 6 in more detail, it emerges that it contains several nodes which

encounter the issue of the inherently qualitative nature of codified themes (e.g. Growth

Strategy, Value Proposition or Mission, Vision & Values), meaning that the similarities

between the main and control sample across the time attributes for those specific nodes

are what should be stressed. The distribution is fairly similar between Historical,

Forward-looking and Non-specific, which does not allow for a lot of concrete interferences.

Hence, the final and most granular level of analysis in Tables 5 and 6 does not seem to

be able to shed more light on the topic.

Approaching the end of the quantitative analysis, two primary conclusions can be drawn.

First of all, analysing voluntary disclosure for the time and depiction attributes initially

established by Beattie et al. (2004) and benchmarking the results against a profitable

group of LEGCs constitutes a useful starting point to derive some initial patterns. This

can be exemplified by inter alia the disproportionally high distribution in the Business

Model Specific KPIs and Alternative Accounting Metrics nodes as well as a particular

non-disclosure behaviour on ‘sensitive’ parameters such as truncation point in time or

cost of equity. Furthermore, the differences between main and control sample are more

distinct for sub-topics and nodes underlying the Financial Information topic than for the

Operating Information topic, which is to be expected given the fact that the companies

in the main and control sample have their profitability as the unique difference, which is

in itself a financial topic. Moreover, doing the Thematic Content Analysis by clustering

for accounting standards, IFRS and U.S.-GAAP, does not yield any major differences in

content as can be seen in Appendix D, which is why the analysis is not driven further.

Secondly, an issue that becomes increasingly apparent throughout the quantitative

analysis is the ‘loss’ of data, a key concern already raised in the literature review by

Healy & Palepu (2001). Although Beattie et al. (2004) started from this limitation and

established a methodology that tries to address this, the aggregation into time and

depiction attributes can be considered an improvement, but is still not sufficient to draw

more concrete conclusions. The most obvious example is the depiction attribute

classification which tends not to be meaningful due to the nature of codified information.

The following sub-section moves the analysis further by doing a qualitative analysis on

the codified themes and comparing the results between the main and control sample.

of 95

Tab

le 6

- D

istrib

utio

n of

Att

ribut

e C

ombi

natio

ns w

ithin

Bus

ines

s M

odel

Nod

es

M

ain

Sam

ple

Con

trol

Sam

ple

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Stra

tegy

Exe

cution

& C

onsist

ency

7530

No

Dat

a25

9A

vaila

ble

Dat

a50

21Q

ualit

ativ

e47

42.0

%42

.0%

10.0

%21

47.6

%47

.6%

4.8%

Qua

ntita

tive

0-

--

0-

--

Both

34.

0%2.

0%-

0-

--

Gro

wth

Str

ateg

y75

30N

o D

ata

339

Ava

ilabl

e D

ata

4221

Qua

litat

ive

394.

8%38

.1%

50.0

%20

14.3

%38

.1%

42.9

%Q

uant

itativ

e0

--

-0

--

-Bo

th3

-7.

1%-

1-

4.8%

-

Val

ue P

ropo

sition

7530

No

Dat

a38

14A

vaila

ble

Dat

a37

16Q

ualit

ativ

e37

29.7

%8.

1%62

.2%

1631

.3%

6.3%

62.5

%Q

uant

itativ

e0

--

-0

--

-Bo

th0

--

-0

--

-

Mission

, V

isio

n &

Val

ues

7530

No

Dat

a44

22A

vaila

ble

Dat

a31

8Q

ualit

ativ

e30

19.4

%-

77.4

%8

--

100.

0%Q

uant

itativ

e0

--

-0

--

-Bo

th1

--

3.2%

0-

--

Tab

le 6

see

subs

eque

nt p

age

for

deta

iled

tabl

e de

scrip

tion

.

of 95

Tab

le 6

- co

ntin

ued

M

ain

Sam

ple

Con

trol

Sam

ple

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

Tot

alH

istor

ical

Forw

ard-

look

ing

Non

-spe

cific

M&

A S

trat

egy

7530

No

Dat

a58

18A

vaila

ble

Dat

a17

12Q

ualit

ativ

e17

17.6

%35

.3%

47.1

%10

41.7

%16

.7%

25.0

%Q

uant

itativ

e0

--

-2

8.3%

8.3%

-Bo

th0

--

-0

--

-

Cus

tom

er I

nfor

mat

ion

7530

No

Dat

a41

17A

vaila

ble

Dat

a34

13Q

ualit

ativ

e21

14.7

%26

.5%

20.6

%7

23.1

%15

.4%

15.4

%Q

uant

itativ

e0

--

-3

15.4

%7.

7%-

Both

1338

.2%

--

323

.1%

--

Em

ploy

ee I

nfor

mat

ion

7530

No

Dat

a29

17A

vaila

ble

Dat

a46

13Q

ualit

ativ

e39

23.9

%26

.1%

34.8

%10

23.1

%23

.1%

30.8

%Q

uant

itativ

e4

4.3%

4.3%

-2

15.4

%-

-Bo

th3

4.3%

2.2%

-1

7.7%

--

R&

D a

nd P

roje

ct A

ctiv

ity

7530

No

Dat

a40

18A

vaila

ble

Dat

a35

12Q

ualit

ativ

e33

37.1

%37

.1%

20.0

%10

25.0

%41

.7%

16.7

%Q

uant

itativ

e0

--

-0

--

-Bo

th2

2.9%

2.9%

-2

-8.

3%8.

3%T

able

6 p

rovi

des t

he d

istrib

utio

n of

att

ribut

e com

bina

tions

with

in th

e eig

ht n

odes

und

erly

ing

the B

usin

ess

Mod

el su

b-to

pic:

Stra

tegy

Exe

cutio

n &

Con

siste

ncy,

Gro

wth

Stra

tegy

, Val

ue P

ropo

sitio

n, M

issio

n, V

ision

& V

alue

s, M

&A

Stra

tegy

, Cus

tom

er I

nfor

mat

ion

, Em

ploy

ee I

nfor

mat

ion

and

R&

D an

d Pr

ojec

t Act

ivity

. Att

ribut

e com

bina

tion

mea

ns ea

ch p

ossib

le co

mbi

natio

n of

the t

hree

tim

e att

ribut

es, i

.e. H

istor

ical

, For

ward

-look

ing

and

Non-

spec

ific,

with

the

four

dep

ictio

n at

trib

utes

, i.e.

No

Data

, Qua

litat

ive,

Qua

ntita

tive

and

Both

(i.e.

Qua

litat

ive

and

Qua

ntita

tive)

. No

Data

illu

stra

tes i

n wh

ich n

odes

vol

unta

ry in

form

atio

n is

not d

isclo

sed

ther

eby

bein

g m

utua

lly ex

clusiv

e with

a

class

ifica

tion

in ei

ther

Qua

litat

ive

or Q

uant

itativ

e or

Bot

h. T

his l

eads

to tw

elve d

iffer

ent c

ombi

natio

n sc

enar

ios (

thre

e tim

e att

ribut

es m

ultip

lied

with

four

dep

ictio

n at

trib

utes

), wh

ereb

y ea

ch fi

rm a

lway

s has

solel

y th

ree c

lass

ifica

tions

per

no

de. A

n ex

ampl

e for

how

to re

ad th

is ta

ble i

s tha

t acr

oss 2

5 m

ain

sam

ple f

irms,

42.0

% o

f the

50

attr

ibut

e cla

ssifi

catio

ns fo

r the

Stra

tegy

, Exe

cutio

n &

Con

siste

ncy

node

(exc

ludi

ng 2

5 No

Dat

a cla

ssifi

catio

ns, i

.e. 7

5 m

inus

25)

are

Q

ualit

ativ

e (d

epict

ion

attr

ibut

e) /

Hist

oric

al (t

ime a

ttrib

ute)

.

of 95

6.2 Qualitative Analysis

Having identified the first patterns and possible data interpretations, the shortcomings

of the quantitative analysis are now overcome with a qualitative analysis. Applying the

methodology established by Gioia et al. (2013), the qualitative analysis for the main

sample yields seven unique Aggregate Dimensions, which can be put into the context of

deduced valuation models: (1) Excess Profitability Generation, (2) Goodwill Duration,

(3) Accounting Conservatism, (4) Internal Goodwill Generation, (5) Pecking Order,

(6) Cost of Capital, and (7) Cost of Equity, and 19 Second Order Themes underlying the

aggregate dimensions respectively. In order to embed and frame the qualitative analysis,

aggregate dimensions are analysed by using the four Forecast Issues which are raised in

the RIV section 3.2 as a theoretical lens.

6.2.1 Regarding Forecast Issue 1

The two aggregate dimensions that emerge from qualitative data analysis are (1) Excess

Profitability Generation and (2) Goodwill Duration, with the first one relating to the

idea that a company only creates value by delivering a book return on owners’ equity in

excess of its required rate of return on owners’ equity, and the second one addressing the

notion that business goodwill/ badwill fades away over time until the company reaches

a competitive equilibrium. Combining the underlying ideas of these two dimensions, they

respectively address the ‘height’ and the ‘length’ of the ‘hump’ necessary to conceptually

explain the market value of LEGCs within a RIV set-up.

(1) Excess Profitability Generation

As explained in section 3.2, only a very strong set of forecast assumptions related to

excess profitability generation capacity can approximate the value that LEGCs have on

the market, given their past and current underlying performance. The stronger these

assumptions are with regard to positive forecasts, the ‘higher' should the ‘hump’ for the

book return on owners’ equity be. Looking at the voluntary disclosure by the sample

companies with regard to (1) Excess Profitability Generation, six main recurring themes

can be observed: (a) Path to Profitability, (b) Operating Leverage, (c) Financial Myopia,

(d) Business Roll-out, (e) Key Stakeholder Management, and (f) ‘Favourable’ Business

Environment.

of 95

Figure 5 summarises the First Order Concepts which lead to these Second Order Themes:

Figure 5 Qualitative analysis summary for Forecast Issue 1 (main sample) using Gioia Methodology


§ Customer-centric strategy to ‘empower’ them§ Invest strongly in brand awareness/ strength§ Strategic refinement/ discontinuations &

engagement in strategic partnerships

§ Invest in leadership & become 1 st choice§ Leadership not clearly defined§ Leadership measured in individual KPIs

(2) Goodwill Duration

(g) Current Competitive Edge

(i) Strategic Initiatives

§ Complexity reduction & ease of use§ Aim to ‘bridge old and new world’ § Service-/ Product-specific competitive edge§ Mission/ Vision linked to multi-faceted edge

(h) Future Leadership

(1) Excess ProfitabilityGeneration

(b) Operating Leverage

(a) Path to Profitability

(c) Financial Myopia

(d) Business Roll-out

(e) Key Stakeholder Management

(f) ‘Favourable’ Business Environment

§ Low entry-barriers due to open software offer§ Fragmented markets which are about to

consolidate; traditional competitors exist§ Newer markets with strong growth potential§ Penetration rate disclosure§ Favourable macro-economic conditions

§ Mention total customers & importance of customer satisfaction (tell anecdotes)

§ Employees as ‘key asset’; plan to hire§ Competitive job market; need for training§ High dependence on key personnel§ Generally high bargaining power of suppliers

with 1 -2 key suppliers; no mitigation efforts

§ Growth strategy focuses on rapid execution§ Geographical expansion key growth factor§ Recent growth strategy adjustments§ Set priorities & clearly aim for profitability§ No M&A strategy; rely more on start-ups/

partnerships and distressed companies§ Mention ‘Go-to-Market’ strategy’ explicitly§ Vague statements about future execution

§ Short-term targets in Business Models KPIs§ General market outlook between 2-3 years§ Short-term revenue guidance, i.e. year-end§ Short-term EPS guidance, i.e. year-end

§ Total variable expenses in % of revenue§ Mention ‘operating leverage’ explicitly§ Disclose contribution margin (no definition)§ Expenses will grow slower than revenues

§ Extensive use of retention-related KPIs§ Often break-even target on EBIT(DA) level§ Heavy use of Gross Margin and EBIT(DA) § Directional revenue guidance§ Revenue growth rates expected to slow down§ Broader guidance for Capex, EPS & FCF

of 95

Focusing on theme (a) to (c), it can be seen that companies try to depict their (a) Path

to Profitability. In fact, they voluntarily disclose information on how their financial

position evolved and will develop presenting the targets they are planning to achieve by

providing for example guidance on traditional line items such as revenue. This shows

that, although LEGCs are high growth companies, which are mostly selling a ‘growth

promise’ to the market, they are aware that they will need to turn profitable at some

point to remain attractive. Nevertheless, the companies remain tentative when it comes

to profitability: the majority of companies in the main sample only discuss profitability

in the sense of reaching the break-even threshold rather than providing future net profits.

What is surprising though is the fact that the information is mostly provided through

the use of Alternative Accounting Metrics (Gross Margin, EBITDA, EBIT or Free Cash

Flow).

“We expect to be breakeven on adjusted EBIDA basis on a monthly level by end of Q4

2018.” (Delivery Hero, Annual Report 2017)

“Furthermore, our goal is to reach breakeven in EBITDA before non-recurring costs

compared to EUR -8.5 million in 2017”

(Shop Apotheke Europe, Quarterly Report 2018)

What is additionally observed is that the companies also discuss how they plan to achieve

their break-even targets: by benefiting from (b) Operating Leverage. Although only some

of the companies explicitly use the term, the idea conveyed is the same as the one

established in academic literature, i.e. operating leverage is the ratio of contribution

margin over operating income, which can be interpreted as a measure of the “sensitivity

of income to changes in sales” (Penman, 2013, p. 409). This ratio is a useful tool to

understand the dependency of the company on fixed costs and whether the profit-

generating ability of a business is positively impacted by a higher scale in revenue. This

is in line with the disclosures explaining how the companies are planning to scale the

business and reduce the relative share of fixed costs as they grow, suggesting to the

readers that profitability will be achieved as they continue to grow. Moreover, this

suggests a high level of understanding of value drivers and a general interest of LEGCs

to help readers, and foremost financial analysts, to do their valuation exercises more

of 95

thoroughly. Nevertheless, none of the companies provides a detailed explanation of how

they calculate their operating leverage or their contribution margin, which makes it

difficult to compare across firms and leaves room for judgment.

“Adjusted EBITDA was $68 million, representing a margin of 18%, up three points

from Q2 of 2017, underscoring that we are driving operating leverage

even as we continue to reinvest in the business.”

(Square, Earnings Call 2018)

The last theme that emerges is (c) Financial Myopia or the idea of short-termism. All

the companies in the main sample provide guidance, targets or goals both for traditional

accounting metrics, such as revenues, expenses or EPS, but also for Business Model

Specific KPIs or Alternative Accounting Metrics. What is noteworthy is the time-frame

over which the guidance is provided, as LEGCs constantly focus on the next one to two

financial years, five years in very rare cases. Revenue and EPS guidance is usually focused

on the end of the year whereas Alternative Accounting Metrics or Business Model Specific

KPIs have two to three years targets. An interesting fact to note is that some companies

are reviewing their guidance policies and no longer guide for very specific metrics; Snap

for example no longer discloses ‘Daily Active Users’ (DAU) guidance.

Overall, companies voluntarily disclose financial information that can help analysts or

other external parties in their valuation exercises. Nevertheless, referring back to the

quantitative analysis, what is important to remember is that the majority of the

information in Additional Financial Data is Historical (53.7%) and within the Forward-

looking time attribute the majority remains Qualitative (47.5%). This shows that

although companies try to address the topics, they do so quite tentatively.

Moving on to (d) Business Roll-out, the following main topics emerge. Firstly, the

business strategies disclosed by LEGCs strongly focus on growth, which can be expected

considering that LEGCs are high growth firms undergoing rapid changes. Nevertheless,

what is important to note is the level at which the discussion is made. In fact,

unprofitable LEGCs usually address their growth strategy on a very high level, using

vague and broad terms and do not provide granular overview or guidance.

of 95

“Our business model focuses on maximizing the lifetime value of a customer

relationship.” (Box, Quartely Report 2018)

Moreover, the companies discuss their growth by describing it as being fast and rapid,

both historically and in the future. Focusing on the historical statements disclosed, what

is observed are recent adjustments to their strategies and execution plans, which could

be either worrisome for external users, as this could signal weak managerial decision-

making abilities, or reassure them, as this could signal managerial steering and the ability

to prioritise in a fast-moving environment. Either way, uncertainty regarding strategic

decisions is to be expected for the young age of LEGCs, which might be characterised

by ‘trial-and-error management’.

Moving on to (e), what is discussed consistently by the sample companies is Key

Stakeholder Management, both in terms of internal and external stakeholders. Employees

are discussed at length in the various financial reports as being a key success factor or a

valuable asset regardless of whether it is senior management or regular workforce. What

is emphasised is the high competition on the job market, which could justify the extensive

disclosure with regard to employee initiatives. This is usually done through employees’

anecdotes or lists of prizes and awards. With regard to external stakeholders, LEGCs

show a high customer focus by showcasing customer satisfaction, in a similar fashion as

for employee, through anecdotes and awards and by highlighting active customer

management initiatives.

Finally, looking into (f), information about ‘Favourable’ Market Conditions are disclosed.

Traditional topics are considered such as competition level, penetration rate, market size,

barriers for new entrants, and future developments. Although the companies compete in

different markets, very similar characteristics are presented: the markets in which LEGCs

operate are characterised by high and increasing competition, high fragmentation

counterbalanced by increasing consolidation, and significant opportunity for growth. Two

interesting recurring topics are that the markets in which the main sample companies

are present are ‘new’ and ‘maturing’ and that competition is centred between ‘old’, i.e.

more traditional, competitors and LEGCs. Moreover, most companies provide

of 95

information about macroeconomic conditions in the various markets in which they are

present. The tone used throughout their statements is positive and tries to convey the

idea that market conditions are ‘favourable’ for the company in order to further support

the assumptions that outside users would have to make in their valuation models.

Comparing these results with the control group (see Appendix E), the following remarks

can be made: first, since it is controlled for profitability, the companies do not mention

(b) Operating Leverage and guidance is much more present with regard to traditional

and Alternative Accounting Metrics within the (a*) Profit Generation theme. Second,

(d) Business Roll-out descriptions are more concrete and are complemented by

justifications. No major differences can be found in the remaining themes, which can be

explained by the overlapping characteristics in the type of firms between the samples.

In summary it can be stated that LEGCs disclose information that can help external

users make assumptions with regard to their (1) Excess Profitability Generation in all

areas of interest (financial, operating, and market environment) which shows both a high

level of valuation literacy and intent to explain their value generation potential.


In addition to the information related to (1) Excess Profitability Generation, deduced

valuation models such as the RIV necessitate an estimate for the truncation point in

time, i.e. the point in time in which business goodwill/ badwill is faded away and the

company reaches a competitive equilibrium. Three main concepts emerging from the

voluntary disclosure by LEGCs that address (2) Goodwill Duration are: (g) Current

Competitive Edge, (h) Future Leadership, and (i) Strategic Initiatives.

What is observed within the data is that companies disclose information which explain

what their respective (g) Current Competitive Edge is based on. Despite the different

wording used to convey uniqueness in their value proposition, similarities across the

sample are present. First of all, the competitive advantage is consistently discussed as

being multi-faceted and having three main characteristics: it is service/ product specific

(i.e. it comes from a unique service/ product that the company is selling), it is a ‘solution’

(i.e. it reduces complexity or increases ease of use for the customers) and it is a bridge

of 95

between the ‘old’ and the ‘new’ world (i.e. it disrupts an established service/ product).

Furthermore, companies also discuss how their brand, mission and vision are key success

factors that sustain their competitive edge. A two-folded comment can be made with

regard to this disclosure: first of all, it is interesting that none of these companies relies

on traditional competitive advantages as discussed in strategy literature (i.e. cost vs.

differentiation (Porter, 1985)), which could be due to the fact that LEGCs operate in

emerging markets which are inherently expected to be more complex (PwC, 2012) and

secondly, voluntarily providing information related to their competitive edge can help

readers assess its strength and hence the durability of the companies’ business goodwill/

badwill. Focusing on forward-looking statements of main sample companies, a clear

theme of (h) Future Leadership emerges. In fact, firms repeatedly state that they strive

to become a ‘leader’. This is also sometimes directly stated in their mission or vision.

Although the way leadership is measured remains very vague and is very inconsistent

across the companies that do inform about the measure, they actively invest in market

leadership as part of their growth strategy and pursue it as their ultimate goal:

“Boozt aims to be the leader in the online apparel industry in the Nordics.”

(Boozt, Annual Report 2017)

Moreover, LEGCs also disclose (i) Strategic Initiatives which will leverage their

competitive edge in order to reach the desired leadership position. The majority of the

initiatives presented by the companies are either customer focused or brand focused. The

first includes statements such as:

“Our growth strategy of build by partner aligns this concept and we are investing to

enable customers to effectively and quickly make decisions and to take actions real

time” (Splunk, Earnings Call 2018)

The latter instead focuses on brand awareness which is supported by disclosures with

regard to increases in sales and marketing expenses planned for the future. What is

interesting to note in this dimension are the strong similarities, despite their ‘uniqueness’,

between all main sample firms’ competitive edge, their future vision of being a ‘leader’

of 95

and the strategic initiatives that will allow them to unlock value generation, which in

turn leads to some general doubt about how many of them will be able to succeed.

Some interesting differences can be found comparing the control sample to the main

sample (see Figure 8 in Appendix E): there is a lack of (h) Future Leadership discussions

which can be explained by the fact that companies in the control sample have a stronger

position in their market and therefore do not feel the need to address the topic. Moreover,

statements with regard to (g) Current Competitive Edge and (i) Strategic Initiatives are

much clearer and more concrete than the ones from the main sample, which is further

strengthened by an apparent strategic consistency.

Overall, the companies in the main sample try to convey the idea that they have a

sustainable competitive edge that will allow them to reach future leadership through

strategic initiatives. This in turn should translate in a RIV set-up into a longer goodwill

duration, thus pushing the truncation point in time further in the future. Combining this

with the findings for (1) Excess Profitability Generation, LEGCs disclose information

that can be used to justify underlying assumptions in a deduced valuation model that

increase both the ‘height’ and the ‘length’ of the ‘hump’. This is less apparent in the

control group, which is to be expected given that the companies are already profitable,

although their statements are overall more concrete.


The two aggregate dimensions that emerge are (3) Accounting Conservatism and

(4) Internal Goodwill Generation.

(3) Accounting Conservatism

Accounting measurement bias, which is a significant parameter in the RIV model, arise

because of the conservative nature of existing accounting standards and the idea of

prudence which allows for neutrality (IASB, 2018, 2.16; FASB, 2018a, BC3.19). Although

companies do not directly discuss accounting measurement bias as defined in the RIV

model (Skogsvik, 2002), what can be found are discussions pertaining to accounting

conservatism. The three main emerging concepts that relate to (3) Accounting

of 95

Conservatism are: (j) Supplemental Performance Reporting, (k) ‘True’ Operational

Performance, and (l) Asset Capitalisation.

Figure 6 Qualitative analysis summary for Forecast Issue 2 (main sample) using Gioia Methodology

As mentioned in the quantitative analysis section, Alternative Accounting Metrics and

Business Model Specific KPIs are the nodes with the most recorded themes, meaning

that LEGCs repeatedly disclose information which are not required or captured by

reporting standards. This is illustrated by the frequent use of KPIs regardless of the topic

discussed (e.g. leadership, customer relationship, expenses or revenues). The strong

reliance on (j) Supplemental Performance Reporting indicators suggests that LEGCs do

not perceive current generally accepted accounting principles to fully capture their

performance drivers, thus having the necessity to thoroughly explain their business to

the market.

In addition to (j) Supplemental Performance Reporting, LEGCs also present a wide

number of adjusted or non-GAAP accounting metrics, including both traditional,

required measures such as net income or expenses and Alternative Accounting Metrics

1 st Order Concepts 2 nd Order Themes AggregateDimensions


(k) ‘True’ Operational Performance

(j) Supplemental Per-formance Reporting

§ Adjusted: Revenues, Expenses, Net Income, Free Cash Flows, Acc. Metrics

§ Non-GAAP: Expenses, Net Income, EPS, Free Cash Flows, Acc. Metrics

§ Constant Currency: Revenues, Acc. Metrics§ Company-specific types of adjustments

§ Highlight their asset-light business models§ Capitalisation rate of intangible assets

(l) Asset Capitalisation

§ KPIs related to customer growth/ retention§ KPIs related to revenue growth/ composition§ Leadership measured in individual KPIs§ Only little KPI-related guidance provided

(4) Internal Goodwill

Generation

(m) Internal Brand Strength

(n) External Brand Strength

§ Initiatives aimed to increase brand awareness§ Key role of brand awareness in success story§ Story-telling & anecdotes of recent successes

§ Actively mention recently won awards§ ‘Name-drop’ customers & outside partners § ‘Name-drop’ indices hinting on customer size§ Report constantly high Net Promoter Scores

of 95

such as EBITDA of Free Cash Flows. The most common adjustments that can be

observed are currency adjustments, i.e. presenting constant currency numbers; stock

compensation adjustments, i.e. excluding stock compensation expenses; non-recurring

expense adjustments, i.e. excluding one-time costs; or other more company-tailored

adjustments (e.g. excluding specific transactions or specific expenses). The idea that can

be discerned through the statements is that current accounting standards do not provide

a (k) ‘True’ Operational Performance for LEGCs because of their conservatism.

Finally, another concept that emerges, but significantly less, is the concept of (l) Asset

Capitalisation. Capitalisation of otherwise expensed items is a way to mitigate the

accounting measurement bias between the booked, historical cost accounting, values and

the ‘fair’, current cost accounting, values. It is a direct component to measure accounting

measurement bias (Runsten, 1998) and it is therefore useful for a RIV model set-up to

have information related to the topic. What is observed are statements of the companies

being ‘asset-light’ and in some cases providing capitalisation rates for their intangibles

assets.

(4) Internal Goodwill Generation

Business goodwill/ badwill is defined as the ability to conduct business projects which

are expected to have a higher internal rate of return than the required rate of return

(Skogsvik, 2002). Goodwill can only be recorded on the balance sheet as an intangible

asset in relation to acquisitions, whereas self-generated goodwill cannot be recognised

(IASB, 2016), which implies that companies can solely communicate the presence of self-

generated goodwill through voluntary disclosure. The two concepts that address Internal

Goodwill Generation are: (m) Internal Brand Strength and (n) External Brand Strength.

As it has been previously noted, LEGCs put a strong emphasis on brand and brand

recognition, which is often discussed as being a key success factor. In fact, companies

engage in recurring disclosure with regard to brand awareness initiatives such as yearly

conferences alongside other big players in the market. This shows the intent to dedicate

internal effort in creating a strong brand. Moreover, companies also extensively discuss

external brand strength and recognition by showcasing awards, ‘Net Promoter Scores’

(NPS) or by providing anecdotes mentioning well recognised companies or persons. This

of 95

has a two-folded interpretation: first it shows that internal efforts are effective and second

it builds credibility in the company by showing how well-respected parties believe in the

company’s potential.

“Yeah. Well, Kirk, I spent a week in D.C. in the quarter. I was in Sydney, where I met

the Prime Minister, I was in UK, we met federal government.”

(ServiceNow, Earnings Call 2018)

Comparing this with the control sample, it appears that the main sample discloses more

information that addresses Forecast Issue 2. In fact, the control sample shows less distinct

evidence of (j) Supplemental Performance Reporting as well less discussions regarding

the role of brand awareness as a key success factor (see Figure 9 in Appendix E).

Summarising, companies indirectly address Forecast Issue 2 in the RIV model by

disclosing information related to both (3) Accounting Conservatism and (4) Internal

Goodwill Generation by respectively providing supplemental or adjusted performance

indicators and communicating the presence of self-generated goodwill. This appears to

be stronger in the main sample than the control group.


With regard to Forecast Issue 3, i.e. expected future growth of owners’ equity to the

horizon point in time, the analysis reveals two aggregate dimensions: (5) Pecking Order

and (6) Cost of Capital. Given the high level of mutual dependency between Forecast

Issue 3 and 4, the identified theme of (q) Leverage Policy is the connecting element,

which is also depicted in Figure 7. A better understanding in the context of the RIV set-

up can be gained through the disclosure of future measures which impact the variables

underlying the Clean Surplus Relation of Accounting as well as voluntary disclosure

about the variables themselves. What is crucial to highlight before, though, is that the

overall amount of first order concepts identified together with the relatively low number

of recorded themes in nodes such as Leverage Ratio or Growth Rate Equity Book Value

(89 and 88 respectively for both samples) clearly makes the point that voluntary

disclosure by LEGCs gives an indication but does not resolve the problems raised in

Forecast Issue 3 and 4 as much as the ones in 1 and 2. Since there are no apparent

of 95

differences between main and control sample, the results for both Forecast Issue 3 and 4

are presented together.

(5) Pecking Order

The first theme of (o) Internal Fund Usage which could be identified across the firms in

the samples, is derived from the Pecking Order Theory of financing, a theoretical

framework, which was first explicitly mentioned by Myers & Majluf (1984). The theory

does not only distinguish between three potential sources of funding but also ranks a

firm’s corporate financing preferences, assuming information asymmetry, in the following

order: internal funds, external debt financing, and external equity financing.

Applying this theoretical background on the context on LEGCs and the established

concept, it becomes apparent throughout the sample firms that the majority does not

only strive for profitability as fast as possible (a key concern for main sample firms), but

also intends to retain and invest the profits they want to generate in the future.

“As we signalled at our half year result, Xero’s organic growth will soon be funded

from our own free cash flows – the next stage of our journey will be one in which we

are self-funding and focused on realising the long-term opportunities

that we are well positioned to secure”

(Xero, Annual Report 2018)

Another observed concept contributing to the theme of (o) Internal Fund Usage is the

overall announcement, for all main sample firms, to not pay dividends in the foreseeable

future given that LEGCs plan to directly reinvest excess profitability. Although this

announcement is arguable in the sense that these firms do not even have the financial

strength to pay out the profits they might generate in the short- or mid-term due to

their accumulated losses, it is still worth highlighting that they rather prefer reinvesting

their own generated funds than relying on external sources (see concepts in (p) Future

Share Issuances, Figure 7).

of 95

“We do not intend to pay cash dividends for the foreseeable future. We have never

declared or paid cash dividends on our capital stock. We currently intend to retain any

future earnings to finance the operation and expansion of our business […]”

(Snap, Quarterly Report 2018)

Overall, it seems like LEGCs do not anticipate having a strong growth of owners’ equity

in the future, especially when it comes to future dividends and share issuances since they

are aiming to finance their operations internally. Nonetheless, the question needs to be

raised of how much external parties can expect (o) Internal Fund Usage to truly happen

for main sample firms since these firms are currently in a loss-making stage at which

they deplete their equity on an annual basis, thus requiring future capital injections.

(6) Cost of Capital – Part I

The aggregate dimension of (6) Cost of Capital encompasses the theme of (q) Leverage

Policy. Specifically, the concepts of capital structure optimisation initiatives, i.e. efforts

to get a capital structure which leads to an optimal/ lower weighted average cost of

capital for the firm, along with the existence of a target leverage ratio should be

highlighted.

“Key focus of the Group will be towards correcting its capital structure – which can

allow for significant profitability to be unlocked.”

(MyBucks, Annual Report 2017)

This is of relevance for Forecast Issue 3, as it allows to determine how the equity share

should increase or decrease annually according to the set leverage policy and how the

underlying variables in the Clean Surplus Relation of Accounting need to be adjusted for

in order to keep the future target leverage ratio stable. This is further strengthened with

disclosure about recently undertaken initiatives with regard to (q) Leverage Policy.

of 95

Figure 7 Qualitative analysis summary for Forecast Issue 3 & 4 (main & control sample) using Gioia Methodology


Continuing the qualitative analysis with Forecast Issue 4, i.e. required rate of return on

owners’ equity to the horizon point in time and in the competitive equilibrium, two

dimensions occur: (6) Cost of Capital and (7) Cost of Equity, with (6) Cost of Capital as

linkage to Forecast Issue 3 (see Figure 7). Information, which could be perceived as

beneficial here would primarily relate to either direct disclosure about a future discount

rate, or input parameters which could be used for the outlined models (e.g. CAPM), or

even data which could help to calibrate the discount rate for the relatively higher

bankruptcy risk of LEGCs.

(6) Cost of Capital – Part II

Continuing with the (6) Cost of Capital dimension introduced beforehand, it can be

stated that none of the LEGCs analysed specifically talks about bankruptcy risk or

reveals data an outside investor could use to assess 𝑝LMNO,0. Solely one company mentions

that it actively manages its entity in order to continue as a going concern.

(7) Cost of Equity(s) Stock Liquidity§ Active interest & disclosure about increased

analyst following§ Increase the share of free-floating shares


(5) Pecking Order

(p) Future Share Issuances

(o) Internal Funds Usage

§ Decisions based on financial health§ Aiming for Profitability as strategic move§ Capex to be funded internally§ No dividends in the foreseeable future

§ Future equity ratio expected to be low§ Less dependent on outside capital injections

(6) Cost of Capital

(q) Leverage Policy

(r) Cost of Debt

§ Disclosure about target leverage ratio§ Capital structure optimisation initiatives

§ Future equity ratio expected to be low

§ Ability to obtain debt funding is impacted§ Increasing cost of debt

of 95

“The Group manages its capital to ensure that entities in the Group will be able to

continue as going concerns while maximising the return to stakeholders through the

optimisation of the debt and equity balance.”

(Takeaway, Annual Report 2017)

The second theme identified under the (6) Cost of Capital dimension is (r) Cost of Debt,

in which some of the firms analysed mention they expect their future cost of debt to

increase, as well as their general ability to obtain debt financing to be negatively affected.

Since these are both first order concepts which indicate that investors do not perceive

the borrowing capacity of LEGCs to be strong right now, it would hardly be an

exaggeration to say that especially some of the main sample firms may encounter

financial distress in the upcoming years.

(7) Cost of Equity

Referring back to what has already been determined within sub-section 6.1.1 of the

quantitative analysis, direct disclosure about an applicable future required rate of return

on owners’ equity cannot be found. The only theme which occurred across some of the

sample firms addresses the topic of a higher future stock liquidity, which the companies

try to improve by increasing the share of free-floating stock as well as by explicitly

informing the market about increased analyst following in the most recent quarters.

“The main purpose of the sale was to improve the tradability of the share, i.e. to

increase the free float and liquidity of the share. […] However, the transaction did not

result in any significant changes to the shareholder structure.”

(Shop Apotheke Europe, Annual Report 2017)

Linking the (7) Cost of Equity dimension to the findings presented in the literature

review, it can be stated that prior findings by Botosan (1997) and Lang & Lundholm

(1993, 1996) about a reduced cost of equity, as well as a lower estimation risk due to

higher financial intermediation play into the voluntary disclosure practice and underline

a potential valuation relevance. The example by Shop Apotheke Europe, however, shows

at the same time that these specific measures undertaken by LEGCs are not always

of 95

soundly executed or create a meaningful impact, which in this example is demonstrated

by the unchanged shareholder structure after the stock issuances.

Summarising the findings from Forecast Issue 3 and 4, there are definitely themes that

can be identified related to the problems addressed. However, as mentioned before, the

very low number of recorded units which pertain to these Forecast Issues makes it hard

to draw meaningful conclusions. Given the expectation that LEGCs would focus on

building credibility around their voluntary disclosure, so that external users integrate

the information provided into their valuation models, the following section is introduced.

6.3 Credibility

As mentioned by Sobel (1985) and Jennings (1987), information is worthless if it is not

credible, which is particularly true in this case: the voluntary information provided by

LEGCs discussed so far will not be used by external readers for their assumptions in

their valuation exercises if they do not believe in it. Given that Credibility is an

encompassing topic that is heavily interlinked with all the findings so far, it is presented

separately in order to provide clarity and a higher-level analysis.

The topic of Credibility has been divided into two sub-topics General Reputation and

Fact Check, which are in turn respectively composed by the Brand Strength and Public

Reputation nodes, and the Degree of Commitment, External Source Reliance, and

Historical Forecast Accuracy nodes. The first sub-topic, inspired by Hoffmann & Fieseler

(2012), can be analysed using both quantitative and qualitative methods given that the

recorded themes are unique and attribute analysis can be done. However, the inherent

nature of the second sub-topic’s nodes makes it meaningless to conduct a quantitative

analysis seeing as an attribute analysis would not yield any results.

With regard to General Reputation, previous research has shown that having a positive

reputation on the market can be financially beneficial for a company (Black et al., 2000),

although it is considered to be the least important element for analysts to form an

impression over a company (Hoffmann & Fieseler, 2012). The quantitative findings of

this thesis suggest that companies might be aware of this contrasting trend which would

of 95

explain why companies voluntarily address the issue but only to a limited extent

(respectively 3.0% and 2.3% of total recorded themes for the main and control sample,

see Table 2). Looking at an attribute combination level, the data is not very insightful,

since it is consistently qualitative with a focus on historical data points, as it could be

expected.

The sub-topic Fact Check is introduced based on Graham et al. (2005) findings as a way

to assess credibility. The reasoning underpinning the topic is that unprofitable LEGCs

cannot rely on a long track record to build credibility around their voluntary disclosure

by consistently meeting guidance or disclosing news in a timely manner (Graham et al.,

2005). They can, however, try to create trustworthiness and convince external users to

rely on the information they provide in their valuation models. Additionally, for

unprofitable LEGCs one could expect them to have a higher interest in building

credibility than profitable companies. The three ways in which this can be done is by:

first, showcasing their Historical Forecast Accuracy so far, which, although limited, can

create a solid foundation for reliability; second, through External Source Reliance for

internal calculations or assumptions; and finally, convey a high Degree of Commitment

by providing clear and measurable guidance, which can be easily back-tested in the

future.

Looking at the first node what is clearly apparent is outperformance; in fact, a majority

of the main sample firms are outperforming previous guidance and are discussing it in a

positive way, creating excitement to showcase high growth and good business, as can be

seen in the following quotes:

“[…] the financial performance has been stronger than we had put in our forecast, so

we’re really happy with that.” (Atlassian, Earnings Call 2018)

“Alex J. Zukin (Analyst, Piper Jaffray & Co.) – ‘[…] Mike, last quarter you mentioned

an outperformance against internal plan. I was curious if you could comment on

performance versus plan this quarter? I mean, how does the pipeline look?’

Michael P. Scarpelli (CFO, ServiceNow, Inc.) – ‘We had a very strong quarter, and

our pipeline looks very good for the second half of the year.’

of 95

John J. Donahoe (CEO, ServiceNow, Inc.) – ‘Which is why we raised guidance.’ ”

(ServiceNow, Earnings Call 2018)

However, it is important to remember that although this could signal positive outlook

for the upcoming months and years, repeated overperformance could have two negative

effects, which are also discussed in Graham et al. (2005): first, it can decrease the

company’s credibility by conveying the idea that management has limited control, and

second, it might mislead external readers into raising expectations for future periods

leading to missing consensus estimates, and ultimately have a negative impact on the

stock price (Skinner & Sloan, 2002). Control group companies are more conservative in

this regard; they usually are in line with previous guidance and meet targets rather than

beat them, which could suggest stronger credibility and better control from management:

“The company’s revenue target was fully met in 2017”

(Zalando, Annual Report 2018)

There is one exception within the control group, Netflix, that is interesting to point out,

considering how contradictive their language is with regard to the issue:

“As a reminder, the quarterly guidance we provide is our actual internal forecast at

the time we report and we strive for accuracy, meaning in some quarters we will be

high and other quarters low relative to our guidance.”

“And as you pointed out, after 4 consecutive quarters of under forecasting the

business, we over forecasted the business. And we strive for accuracy.”

“And Todd, we’ve seen this movie of Q2 shortfall before about 2 years ago in 2016,

and we never did find the explanation of that other than there’s some lumpiness in

the business and continue to perform after that.”

(Netflix, Earnings Call 2018)

External source reliance is not frequent across either the main or the control sample, and

the few companies that do disclose sources mostly rely on GDP estimates from

of 95

governmental bodies to estimate current and future addressable market sizes. One

surprising finding is that some companies disclose the fact that they have stopped relying

on external data to measure internal KPIs as they perceive internal calculations to be

more accurate.

“In the past we have relied on third-party analytics providers to calculate our metrics,

but today we rely primarily on our analytics platform that we developed and operate”

(Snap, Annual Report 2017)

Although this might be true, investors and analysts might be less inclined to trust these

metrics, since transparency decreases and there is more room for manipulation to

showcase positive results.

Regarding Degree of Commitment, a considerable difference can be observed between

the control group and the main sample. First of all, the control group discloses

substantially more commitments statements, i.e. statements in which they set

expectations or targets, seeing as they repeat their expectation statements multiple times

within and across documents, which is done to a lesser extent by the main sample.

Moreover, there is a clear dominance of quantitative information in the control group,

since they provide numerical targets and goals which are easy to back-test and verify,

whereas the main sample usually only commits directionally by specifying whether they

expect an increase or decrease without quantifying the change. Additionally, in cases

where LEGCs in the main sample provide quantitative targets, they often do so through

the use of ranges, and although this can be found in the control group too, the size of

the ranges is larger for the main sample than the control group.

“PayPal expects revenue to grow 17 - 19% at current spot rates and 16 - 18% on an

FX-neutral basis, to a range of $15.30 - $15.50 billion. As previously disclosed, full

year 2018 revenue guidance includes an expected impact related to the sale of U.S.

consumer credit receivables to Synchrony Financial of approximately 3.5 percentage

points for full year 2018.”

(PayPal, Press Release 2018)

of 95

“We will also make efforts to significantly grow LINE MUSIC.”

(Line, Earnings Call 2018)

Summarising, although unprofitable LEGCs are trying to build a foundation for

credibility by actively making credibility-building statements addressing Historical

Forecast Accuracy, and by providing guidance, continuing overperformance and a

majority of directional guidance could suggest that external users might be less inclined

to rely on voluntarily disclosed information in the long run. Moreover, the clear

distinction that can be observed between the main and control group suggesting that a

relation might exist between profitability and credibility.

The analyses yield insights on the voluntary disclosure by LEGCs by using the RIV’s

underlying Forecast Issues as a theoretical lens. However, to answer the research

question, it is now crucial to verify if the Forecast Issues are solved. The following section

will therefore discuss the matter, synthesising the findings from the analyses. Moreover,

suggestions as of why management of LEGCs engages in voluntary disclosure, or rather

non-disclosure, practices are additionally outlined in the discussion.

7 Discussion & Limitations

7.1 Voluntary Disclosure – A Solution to the Forecast Issues?

In order to understand how voluntary disclosure can contribute to explain the valuation

gap between fundamental accounting numbers and the stock market price, it is crucial

to understand whether the information disclosed solves the Forecast Issues that underpin

a RIV set-up. What is meant by solving the Forecast Issues is assessing whether, and

how, the observed LEGCs’ voluntary disclosure contains information that can support

the strong set of assumptions necessary to justify their market value.

From the analyses conducted, the following conclusions with regard to the four Forecast

Issues (Skogsvik, 2002, 2006) presented in section 3.2 can be drawn. First of all, LEGCs

do address Forecast Issue 1, i.e. the expected future book return on owners’ equity until

the truncation point in time, by explaining their future value generation potential both

financially and operationally, as well as depicting favourable market conditions, which

of 95

can be used by external users to make forecasts of the expected future book return on

owners’ equity. With regard to Forecast Issue 2, i.e. the expected relative goodwill/

badwill at the truncation point in time, although not discussed directly, LEGCs convey

the idea that existing accounting reporting standards are insufficient to showcase their

value drivers, suggesting a positive and possibly high accounting measurement bias at

the truncation point in time. Forecast Issue 3 and 4, respectively expected future growth

of owners’ equity at the truncation point in time and required rate of return on owners’

equity, are not discussed in detail although some information could be of use to analysts.

Despite the fact that the first two Forecast Issues are addressed, it is important to assess

how and to what extent, before concluding that LEGCs provide information that solves

them. Synthesising the content shows that companies in the main sample do not provide

a large amount of quantitative information which can be directly inserted in the

assumptions of the valuation models, but rather present information through qualitative

statements that explain the future development of the company operationally and

financially. A primary example is the discussion related to their (g) Current Competitive

Edge that was identified in the analysis. The underlying idea which could justify

disclosure of such topics is that LEGCs want to convince external readers by informing

them about their sustainable competitive edge, unique value proposition or achievable

growth strategy, ultimately enabling them to better resist competitive threats from either

existing or potential new incumbents. This would allow business goodwill to last longer

in a valuation model setup. Nevertheless, the statements disclosed are often general or

vague, meaning that they are not always company-specific. These two issues make it

more complicated for external users to directly insert the information in their valuation

models, but also harder to draw insights on the companies’ value drivers without

substantially interpreting and processing the data first.

Additionally, LEGCs exhibit a high level of (c) Financial Myopia, meaning that the

information they voluntarily disclose consistently focuses on a year-end to two-year

period, which in several cases only guides until the break-even event, thus being

insufficient to fully shed light on the forecast period until the truncation point in time.

This (c) Financial Myopia is surprising, given the valuation setup, since it is a necessary

precondition for excess profitability to not only be consistently high, but to also be

of 95

sufficiently long-lasting to conceptually justify the current difference between reported

numbers and market value. This is particularly true for LEGCs who are at an unstable

stage of their business life-cycle and are rather selling a growth story to the market for

which a ‘hump-shaped’ forecast is conceptually necessary to justify their value on the

market.

Finally, given the expectation that LEGCs in the main sample would focus on building

credibility around their voluntary disclosure, so that external users integrate the

information provided into their valuation models, credibility statements appear to be

surprisingly moderate, especially compared to the control group. An example that

strongly portrays this idea is the difference in statements with regard to Degree of

Commitment, where the main sample gives directional guidance and broader numerical

ranges than the control group. This could diminish the extent to which one would expect

external users to rely on the information provided to feed the assumptions of deduced

valuation models, such as RIV.

7.2 Non-Disclosure – ‘Window Dressing’ or Too Sensitive to Disclose?

Conducting a study about voluntary disclosure additionally requires investigating the

specific areas of non-disclosure. Since this thesis draws from the idea that firms use

voluntary disclosure to overcome information asymmetries (Verrecchia, 2001) between

management and investors/ external parties about future cash-flows as well as

stewardship (IASB, 2018; FASB, 2018a), it becomes apparent that LEGCs seem to make

strategic considerations when they reveal voluntary information to the market. Although

discretionary research (Verrecchia, 2001) is not a focus for this thesis, it is of interest to

discuss the motives that LEGCs could have in maintaining information asymmetry in

specific areas. This sub-section illustrates two popular reasons for this pattern in the

context of financial disclosure.

To begin with, it is necessary to first have a closer look at the research area of impression

management. Reviving the impression management strategies summarised by Merkl-

Davies & Brennan (2007), special attention should be paid to Choice of Earnings

of 95

Number, a strategy serving the managerial motive of concealment. By selectively

disclosing a wide range of numerical performance measures such self-constructed Business

Model Specific KPIs as well as Alternative Accounting Metrics, which do not go further

down the income statement than EBIT(DA), LEGCs seem to actively try to create a

bias within their financial reporting. Since bottom-line measures such as book return on

owners’ equity would definitely not favourably portray the companies given their current

stage, it can be seen that non-disclosure in traditional financial metrics is exercised fairly

often, thus leading to the impression of ‘window dressing’. This behaviour can also be

observed in disclosures related to (3) Accounting Conservatism and (a) Path to

Profitability. Regarding the first, although companies are trying to convey the idea that

adjusted measures allow a more fair and comparable representation of the company, one

cannot eliminate the possibility that management is engaging in some form of impression

management by manipulating or choosing the numbers in order to depict a more positive

image of the company (Merkl-Davies & Brennan, 2007). With regard to the second, as

discussed in Graham et al. (2005), not meeting guidance is badly perceived by the market

and can have repercussions on both management’s reputation and the stock price

(Skinner & Sloan, 2002). Therefore, it is a reasonable consideration for companies to

limit the disclosure of clear quantitative guidance unless certain. These observations

furthermore strengthen the evidence by Black et al., (2000) and McGuire et al. (1990)

that firm reputation has significant value-relevance, thereby giving LEGCs an incentive

to engage in impression management practices. Ultimately, impression management is a

key concern for financial statement users, who should consider such practices when

running a valuation exercise and not fall into this trap.

Besides the managerial motive of concealment, a second discussion about the

fundamental costs which occur by carrying out voluntary disclosure is necessary. Linking

back to the costs outlined by Graham et al. (2005), a primary focus is laid on the topic

of proprietary cost, since this is a crucial point for companies with firm characteristics of

LEGCs. Given the dynamic market environment and the recent listing of these firms,

the competitive advantage could be decisively harmed if the company management were

to disclose sensitive information competitors could benefit from. A key advantage of the

quantitative analysis is that it strongly hints to sub-topics in which either no information

is disclosed (e.g. Truncation Point in Time or Cost of Equity) or the attribute

of 95

combination Qualitative / Forward-looking (e.g. R&D and Project Activity) often

occurs. Complementing this pattern with the qualitative analysis, it is possible that

proprietary costs constrict the voluntary disclosure strategy by, for example, influencing

company management to be extremely careful in terms of content choice and vague in

terms of tone about future R&D efforts, where they mainly disclose that they expect the

costs to increase, but do not inform about the projects they will invest in. This can

additionally explain the limited quantitative guidance with regard to financial metrics

provided by LEGCs in the main sample and the interruption of some KPI guidance.

Hence, showcasing a significant trade-off between informing the market while, at the

same time, not jeopardising the firm value, ultimately leading to an intended information

asymmetry.

Although within the established research design it is impossible to lean towards one of

the two non-disclosure reasons outlined, it becomes apparent that there is a fine line

between not disclosing information due to concealment reasons and due to proprietary

cost reasons. While information is not presented in the first one in order to safeguard the

current market value based on flawed impression, the second one aims to safeguard the

current market value by protecting for example the firm’s intellectual property.

Management may apply an argumentation from either one of the above reasons

interchangeably and assessing this with more details would require an inside study about

disclosure strategy.

A final remark about potential non-disclosure with regard to financial guidance is that

it may also relate to the fact that the information required simply does not exist given

the business life-cycle stage and the financially unstable situation of LEGCs, making it

hard for management to provide voluntary disclosure about certain topics. This could

additionally justify the differences observed between main and control sample with

regard to the amount of quantitative data related to deduced valuation models, the

presence of Quantitative / Forward-looking (j) Supplemental Performance Reporting,

and the Degree of Commitment disclosed.

of 95

7.3 Limitations

To provide an overview of the main limitations of this thesis, three aspects should be

highlighted. First of all, it is evident that a purposive sampling method with 25 LEGCs

representing the main sample limits the ability to draw generalisations with regard to

the entire population of LEGCs. However, it is important to restate that the overall

purpose of this thesis is not to generate conclusions for a specific population, but rather

to understand how voluntary disclosure efforts contribute to explain the high market

price of rapidly growing and unprofitable companies. In addition, it is crucial to highlight

that the population of firms with the outlined criteria is inherently limited, as the

emergence of LEGCs constitutes a recent phenomenon.

Secondly, controlling for profitability in the quantitative analysis does not yield

sufficiently significant insights. Potentially, more valuable results could be generated by

controlling for different characteristics, e.g. disclosure practice by stable and mature firm

in a related industry. Besides the scope of this thesis, which did not allow for a further

investigation into this area, the point needs to be brought up that the general disclosure

practice of these firms might be much more established, making it ultimately easier to

draft more comprehensive reports and consequently more data points related to

voluntary disclosure.

Thirdly, an issue that arises with the application of qualitative research is the concern

of subjectivity. Although this is a fair objection, it becomes apparent throughout this

thesis that quantitative analysis as suggested by Beattie et al. (2004) or alternatives such

as ‘key word search’ by Athanasakou et al. (2018) fail to broach the issue of a detailed

content of the revealed information, demonstrating the necessity of qualitative analysis.

of 95

8 Conclusion & Implications

8.1 Conclusion

This thesis investigated the contribution of voluntary disclosure to conceptually explain

the valuation gap between fundamental accounting numbers and the stock market price

of Listed Emerging Growth Companies within a RIV set-up using mixed methodology.

The data are analysed using four Forecast Issues which occur undertaking a Residual

Income Valuation as the theoretical lens to understand how and to what extent LEGCs

address these issues, ultimately aiming to provide additional information for external

users to formulate their model assumptions. Having conducted an analysis on the main

sample of 25 unprofitable LEGCs and contrasting it with a control sample of 10 profitable

LEGCs, the findings suggest that voluntary disclosure contributes to explain the market

value of LEGCs through Forecast Issue 1 and 2 on a short to medium term.

Forecast Issue 1, i.e. expected future book return on owners’ equity to the horizon point

in time and Forecast Issue 2, i.e. expected relative goodwill/ badwill of owners’ equity

at some ‘appropriately’ chosen horizon point in time, are addressed with voluntary

disclosure by the firms investigated. The first is inter alia discussed through the theme

of (b) Operating Leverage, however, only on a short to medium term, demonstrating

(c) Financial Myopia. The second is amongst others illustrated by (j) Supplemental

Performance Reporting using Business Model Specific KPIs as well as Alternative

Accounting Metrics. While the mentioned data points do supplement the valuation

exercise for outside investors, it is apparent that the vast majority cannot directly flow

into a RIV model, meaning that outside parties, such as financial analysts, have to

substantially process and work with the data, ultimately leaving room for interpretation

to external users. Combining this with an assessment of the overall degree of credibility

of the information disclosed, i.e. substantiating the presence of credibility-building

statements such as Historical Forecast Accuracy, suggests that external users might be

less inclined to rely on the voluntarily disclosed information, contingent on the fact that

they question the information presented in the first place.

Lastly, this thesis further finds evidence about the existence of specific areas of either

non-disclosure or highly vague and qualitative, forward-looking statements, relating to

of 95

direct model variables such as truncation point in time but also upcoming R&D projects.

Although it is beyond the scope of this thesis, this may hint to a wider debate in the

area of discretionary research dealing with proprietary cost considerations or impression

management strategies. Overall, the outlined findings along with the above-mentioned

debate strongly indicate the relevance of re-entering into the topic of voluntary disclosure

and serve as a springboard for future research.

8.2 Contributions to Literature

This thesis contributes to the existing disclosure literature in three different ways. Firstly,

prior literature has relied to a large extent on self-constructed disclosure measures or on

AIMR and FAF Reports, which, besides the discontinuation of the reports in 1997, have

both been criticised because of endogeneity issues and over-aggregation. Furthermore,

the methodology of Attribute Analysis outlined by Beattie et al. (2004), is considered a

development compared to the previously existing Thematic Content Analysis by

overcoming the aforementioned downside of over-aggregation, but has never really been

applied in practice. The contribution of this thesis lies in the application of a mixed

method approach starting off from a high-level Thematic Content Analysis to a more

thorough Attribute Analysis suggested by Beattie et al. (2004) and finishing off with the

most granular way of looking at the data, doing a qualitative analysis in line with the

method described by Gioia et al. (2013). The authors strongly believe that the

thoroughness of a mixed method is a necessity to conduct meaningful voluntary

disclosure studies, although this demands a high degree of manual work making it

complicated to replicate and implement on a large scale.

Secondly, looking at the topic of value relevance by conceptually embedding it into the

set-up of the RIV model, and thereby approaching it with a deduced valuation model

angle instead of a capital markets one, gives this thesis an edge compared to prior studies

undertaken in the area of association-based literature by coming from the fundamental

accounting numbers. Although coming from this perspective is, in itself, not completely

novel (Gray & Skogsvik, 2004), it remains a rather untapped research area in which the

authors are able to make first contributions.

of 95

Thirdly, and most importantly, this thesis contributes to existing literature by focusing

on a new type of firms, Listed Emerging Growth Companies, which could be expected to

become more popular in the future. With today’s rapidly changing market environment

and the current IPO trends introduced at the beginning, it seems crucial to study this

recent phenomenon in order to understand if previous findings in the voluntary disclosure

literature still hold. The identified core themes of LEGCs such as (b) Operating Leverage,

(c) Financial Myopia, or (j) Supplemental Performance Reporting represent a valuable

starting point for future research.

8.3 Areas of Further Research

Referring back to Ritter’s statistics (2018) that 83.0% of the technology firms doing an

IPO in 2017 were unprofitable, it can be observed that the presence of LEGCs has become

an increasing trend over the last seven years. By studying LEGCs, the key ambition of

this thesis is to encourage further research in the area of value relevance of voluntary

disclosure. Despite the limited scope, relevant and interesting findings were generated

encouraging further research which either build upon or complement this thesis’ findings.

To begin with, while the overall research approach of this thesis is highly conceptual, a

practical application could be of interest where, for example, the market value of owners’

equity is reverse engineered, showing how much of the market value can be numerically

explained by mandatory information complemented with statistically well-behaved

assumptions going forward compared to mandatory information complemented with a

forecast based on voluntarily disclosed information.

In addition, replicating the concept of this thesis on a larger scale using, for example, a

random sampling technique, could help draw relevant and generalisable conclusions for

a population of firms, even though this would require to numerically capture the

differences in quality and content of voluntary disclosure which can be uncovered through

a qualitative analysis.

Studies could also further engage in the discussion raised with regard to proprietary cost

considerations or impression management strategies, by investigating the factors

of 95

influencing voluntary disclosure practices in a firm. This could be done by interviewing

or surveying teams (e.g. in the financial reporting department) and key personnel (e.g.

CFOs) involved in the drafting of financial disclosure documents (similar to Graham et

al., 2005)

Looking closer into the topic of investor clientele is a final suggestion for future research.

This thesis implicitly assumes that the market value of LEGCs is a consequence of

external investors who act in a rational way when it comes to capital allocation decisions,

ultimately supporting a potential value relevance of voluntary disclosure. However, if

external investors belong to the category of ‘gamblers’, meaning that they actively enter

a bet on the ‘growth promise’ of a specific firm based on their ‘gut feeling’ rather than

processed information, the value relevance of voluntary disclosure for LEGCs could be

strongly questioned.

Finally, it is important to point out that further research on voluntary disclosure could

support financial reporting bodies such as the IASB and FASB to revise disclosure

requirements in order to focus on more value-relevant information in mandatory reports.

of 95

9 References

Adelberg, A. H. (1979). Narrative Disclosures Contained in Financial Reports: Means of Communication or Manipulation? Accounting and Business Research, 9(35), 179–190.

Akerlof, G. A. (1970). The Market for “Lemons”: Quality Uncertainty and the Market Mechanism. The Quarterly Journal of Economics, 84(3), 488–500.

ACCA (Association of Chartered Certified Accountants), (2012). Re-assessing the value of corporate reporting. Accountancy Futures.

Athanasakou, V., El-Haj, M., Rayson, P., Walker, M., & Young, S. (2018). Management Commentary Articulating Strategy and Business Model: Measurement and Impact.

Balakrishnan, K., Billings, M. B., Kelly, B., & Ljungqvist, A. (2014). Shaping Liquidity: On the Causal Effects of Voluntary Disclosure. The Journal of Finance, 69(5), 2237–2278.

Beattie, V., McInnes, B., & Fearnley, S. (2004). A methodology for analysing and evaluating narratives in annual reports: A comprehensive descriptive profile and metrics for disclosure quality attributes. Accounting Forum.

Bennett, N., & Lemoine, G. J. (2014). What a difference a word makes: Understanding threats to performance in a VUCA world. Business Horizons, 57(3), 311–317.

Black, E. L., Carnes, T. A., & Richardson, V. J. (2000). The Market Valuation of Corporate Reputation. Corporate Reputation Review, 3(1), 31–42.

Botosan, C. a. (1997). Disclosure Level and the Cost of Equity Capital. The Accounting Review, 72(3), 323–349.

Brown, S., Hillegeist, S. A., & Lo, K. (2004). Conference calls and information asymmetry. Journal of Accounting and Economics, 37(3), 343–366.

Bryman, A., & Bell, E., (2011). Business Research. Methods. Oxford University Press (3rd ed.).

Paredes, T. A., (2013). Remarks at The SEC Speaks in 2013. Retrieved September 28, 2018, from https://www.sec.gov/news/speech/2013-spch022213taphtm#P35_8886

Core, J. E. (2001). A review of the empirical disclosure literature: Discussion. Journal of Accounting and Economics, 31(1–3), 441–456.

Courtis, J. K. (2004). Corporate report obfuscation: Artefact or phenomenon? British Accounting Review, 36(3), 291–312.

of 95

Damodaran, A. (2009). Valuing Young, Start-Up and Growth Companies: Estimation Issues and Valuation Challenges. SSRN Eletronic Journal, 1–67. 7

Deloitte. (2017). Annual report insights 2017 Surveying FTSE reporting.

Dhaliwal, D. S., Li, O. Z., Tsang, A., & Yang, Y. G. (2011). Voluntary nonfinancial disclosure and the cost of equity capital: The initiation of corporate social responsibility reporting. The Accounting Review, 86(1), 59–100.

Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure, liquidity, and the cost of capital. The Journal of Finance, 46(4), 1325–1359.

Driebusch, C., & Farrell, M. (2018, October 1). IPO Market Has Never Been This Forgiving to Money-Losing Firms; Money-losing companies are going public at a record rate as investors hunger for new issues. Wall Street Journal. Retrieved from https://search.proquest.com/docview/2114604393?accountid=39039

Edwards, E., & Bell, P. W. (1961). The Theory and Measurement of Business Income. Garland Publishing Inc.

EU (European Parliament and European Council) Directive 2014/95/EU amending Directive 2013/34/EU, Disclosure of non-financial and diversity information by certain large undertakings and groups, OJ L 330, 15.11.2014, p. 1–9.

EY. (2014). Disclosure effectiveness: What investors, company executives and other stakeholders are saying.

EY. (2018). IFRS Core Tools - International GAAP Disclosure Checklist.

Fama, E., & French, K. (1993). Common Risk Factors in the Returns on Stocks and Bonds. Journal of Financial Economics, 33, 3–56.

Farris, P. W., Bendle, N., Pfeifer, P., & Reibstein, D. (2010). Marketing metrics: The definitive guide to measuring marketing performance. Pearson Education. 2nd Edition.

FASB (Financial Accounting Standards Board). (2018a). Conceptual Framework for Financial Reporting. Statement of Financial Accounting Concepts No. 8.

FASB (Financial Accounting Standards Board). (2018b). FASB Accounting Standards Codification. Retrieved from https://www.fasb.org/

Francis, J., Philbrick, D., & Schipper, K. (1994). Shareholder Litigation and Corporate Disclosures. Journal of Accounting Research, 32(2), 137–164.

Francis, J., Douglas Hanna, J., & Philbrick, D. R. (1997). Management communications with securities analysts. Journal of Accounting and Economics (Vol. 24).

of 95

Francis, J., Nanda D., & Olsson P. (2008). Voluntary Disclosure, Earnings Quality and Cost of Capital. Journal of Accounting Research, 46(1), 53–99.

Gioia, D. A., Corley, K. G., & Hamilton, A. L. (2013). Seeking Qualitative Rigor in Inductive Research. Organizational Research Methods, 16(1), 15–31.

Gordon, M. J., & Shapiro, E. (1956). Capital Equipment Analysis: The Required Rate of Profit. Management Science, 3(1), 102–110.

Graham, J. R., Harvey, C. R., & Rajgopal, S. (2005). The economic implications of corporate financial reporting. Journal of Accounting and Economics.

Gray, S. J., Skogsvik, K. (2004). Voluntary disclosures of quoted pharmaceutical companies in Sweden and the UK: The development over the period 1984-98. European accounting Review, 13(4), 787-805.

Gu, F., & Li, J. Q. (2007). The credibility of voluntary disclosure and insider stock transactions. Journal of Accounting Research, 45(4), 771–810.

Healy, P. M., & Palepu, K. G. (1993). The Effect of Firms’ Financial Disclosure Strategies on Stock Prices. Accounting Horizons, 7(1), 1–11.

Healy, P. M., Hutton, A. P., & Palepu, K. G. (1999). Stock Performance and Intermediation Changes Surrounding Sustained Increases in Disclosure. Contemporary Accounting Research, 16(3), 485–520.

Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature. Journal of Accountancy and Research, 31(1–3), 405–440.

Hoffmann, C., & Fieseler, C. (2012). Investor relations beyond financials: Non-financial factors and capital market image building. Corporate Communications, 17(2), 138–155.

IASB (International Accounting Standards Board). (2016). A Guide through IFRS Standards.

IASB (International Accounting Standards Board). (2018). Conceptual Framework.

Jennings, R. (1987). Unsystematic Security Price Movements, Management Earnings Forecasts, and Revisions in Consensus Analyst Earnings. Journal of Accounting Research, 25(1), 90–110.

Johnson, R. B., & Onwuegbuzie, A. J. (2004). Mixed Methods Research: A Research Paradigm Whose Time Has Come. Educational Researcher, 33(7), 14–26.

of 95

Jones, M. M. J., & Shoemaker, P. A. (1994). Accounting narratives: A review of empirical studies of content and readability. Journal of Accounting Literature, 13, 142.

Kaplan, R. S., & Norton, D. P. (2008). Mastering the management system. Harvard Business Review, 86(1), 62.

Koller, T., Goedhart, M., & Wessels, D. (2015). Valuation - Measuring and Managing the Value of Companies. John Wiley and Sons. 6th Edition.

KPMG. (2011). Disclosure overload and complexity: hidden in plain sight.

KPMG. (2016). Room for improvement: The KPMG Survey of Business Reporting. (Vol. 2).

KPMG. (2017a). The road ahead: The KPMG Survey of Corporate Responsibility Reporting. (Vol. 10).

KPMG. (2017b). Disclosure Checklist - Guide to annual financial statements IFRS.

Lang, M., & Lundholm, R. (1993). Cross-Sectional Determinants of Analyst Ratings of Corporate Disclosures. Journal of Accounting Research, 31(2), 246–271.

Lang, M., & Lundholm, R. (1996). Corporate disclosure policy and analyst behavior. Accounting Review, 71(4), 467–492.

Lang, M., & Stice-Lawrence, L. (2015). Textual analysis and international financial reporting: Large sample evidence. Journal of Accounting and Economics, 60(2.3), 110–135.

Leary, M. R., & Kowalski, R. M. (1990). Impression Management: A Literature Review and Two-Component Model. Psychological Bulletin, 107(1), 34–47.

Lintner, J. (1965). The Valuation of Risk Assets and the Selection of Risky Investments in Stock Portfolios and Capital Budgets. The Review of Economics and Statistics, 47(1), 1–13.

McGuire, J. B., Schneeweis, T., & Branch, B. (1990). Perceptions of Firm Quality: A Cause or Result of Firm Performance. Journal of Management, 16(1), 167–180.

Merkl-Davies, D., & Brennan, N. (2007). Discretionary Disclosure Strategies in Corporate Narratives: Incremental Information or Impression Management? Journal of Accounting Literature, 27, 116–196.

Michelon, G., Pilonato, S., & Ricceri, F. (2015). CSR reporting practices and the quality of disclosure: An empirical analysis. Critical Perspectives on Accounting, 33, 59–78.

Morris, M., Schindehutte, M., & Allen, J. (2005). The entrepreneur’s business model: Toward a unified perspective. Journal of Business Research, 58(6), 726–735.

of 95

Mossin, J. (1966). Equilibrium in a Capital Asset Market. Econometrica, 34(4), 1–768.

Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics, 13, 187–221.

O’Cathain, A., Murphy, E., & Nicholl, J. (2007). Integration and Publications as Indicators of “Yield” From Mixed Methods Studies. Journal of Mixed Methods Research, 1(2), 147–163.

Onwuegbuzie, A. J., & Johnson, R. B. (2006). The Validity Issue in Mixed Research. Research in the Schools, 13(1), 48–63.

Penman, S. H. (2013). Financial Statement Analysis and Security Valuation. McGraw-Hill. 6th Edition.

Porter, M. E. (1979). How Competitive Forces Shape Srategy. Harvard Business Review, 102–117.

Porter, M. E. (1985). Competitive Advantage: The Core Concepts. In Competitive Advantage: Creating and sustaining superior performance (pp. 11–25). New York: The Free Press.

Porter, M. E. (2008). The five competitive forces that shape strategy. Harvard Business Review, 86(1), 78–93.

Preinreich, G. A. D. (1938). Annual Survey of Economic Theory: The Theory of Depreciation. Econometrica, 6(3), 219–241.

PwC. (2012). Emerging industries. European Union.

PwC. (2017). Value IFRS Plc - Illustrative IFRS consolidated financial statements.

PwC. (2018). IFRS and US GAAP: similarities and differences.

Ritter, J. R. (2018). Initial Public Offerings: Updated Statistics.

Runsten, M. (1998). The Association between Accounting Information and Stock Prices – Model development and empirical tests based on Swedish data. Stockholm School of Economics.

Rutherford, B. A. (2003). Obfuscation, textual complexity and the role of regulated narrative accounting disclosure in corporate governance. Journal of Management and Governance, 7(2), 187–210.

Schick, A. G., Gordon, L. A., & Haka, S. (1990). Information overload: A temporal approach. Accounting, Organizations and Society, 15(3), 199–220.

of 95

Schoenfeld, J. (2017). The effect of voluntary disclosure on stock liquidity: New evidence from index funds. Journal of Accounting and Economics, 63(1), 51–74.

SEC. (2017). Financial Reporting Manual.

Sharpe, W. F. (1964). Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk. The Journal of Finance, 19(3), 425–442.

Skinner, D. J., & Sloan, R. G. (2002). Earnings Surprises, Growth Expectations, and Stock Returns: Don’t Let an Earnings Torpedo Sink Your Portfolio. Review of Accounting Studies, 7(2–3), 289–312.

Skogsvik, K. (1998). Conservative accounting principles, equity valuation and the importance of voluntary disclosures. British Accounting Review, 30(4), 361–381.

Skogsvik, K. (2002). A Tutorial on Residual Income Valuation and Value Added Valuation second revision). SSE/EFI Working Paper Series in Business Administration, 1999(10).

Skogsvik, K. (2006). Probabilistic Business Failure Prediction in Discounted Cash Flow Bond and Equity Valuation. SSE/EFI Working Paper Series in Business Administration.

Sobel, J. (1985). A Theory of Credibility. The Review of Economic Studies, 52(4), 557–573.

Verrecchia, R. E. (2001). Essays on disclosure. Journal of Accounting and Economics, 32(1–3), 97–180.

Wagenhofer, A. (1990). Voluntary disclosure with a strategic opponent. Journal of Accounting and Economics, 12(4), 341–363.

of 95

10 Appendices

10.1 Appendix A – EGC & SME Definition

Emerging Growth Companies (EGC)

Small- and Medium-sized Entities (SME)

Defined by SEC: https://www.sec.gov/smallbusiness/goingpublic/EGC

IASB: https://www.ifrs.org/supporting-implementation/supporting-materials-for-the-ifrs-for-smes/

Definition A company qualifies as an EGC if it has total annual gross revenues of less than $1.07 billion during its most recently completed fiscal year and, as of December 8, 2011, had not sold common equity securities under a registration statement. A company continues to be an emerging growth company for the first five fiscal years after it completes an IPO, unless one of the following occurs: (a) its total annual gross revenues

are $1.07 billion or more, (b) it has issued more than $1

billion in non-convertible debt in the past three years, or

(c) it becomes a “large accelerated filer,” as defined in Exchange Act Rule 12b-2.

SMEs are entities that: (a) do not have public accountability; and (b) publish general purpose financial statements for external users. Examples of external users include owners who are not involved in managing the business, existing and potential creditors, and credit rating agencies. An entity has public accountability if: (a) its debt or equity instruments

are traded in a public market or it is in the process of issuing such instruments for trading in a public market (a domestic or foreign stock exchange or an over-the-counter market, including local and regional markets), or

(b) it holds assets in a fiduciary capacity for a broad group of outsiders as one of its primary businesses (most banks, credit unions, insurance companies, securities brokers/dealers, mutual funds and investment banks would meet this second criterion).

of 95

continued Emerging Growth Companies

(EGC) Small- and Medium-sized

Entities (SME) Example of Disclosure Differences

EGCs are permitted: § to include less extensive

narrative disclosure than required of other reporting companies, particularly in the description of executive compensation

§ to provide audited financial statements for two fiscal years, in contrast to other reporting companies, which must provide audited financial statements for three fiscal years

§ not to provide an auditor attestation of internal control over financial reporting under Sarbanes-Oxley Act Section 404(b)

§ to defer complying with certain changes in accounting standards and

§ to use test-the-waters communications with qualified institutional buyers and institutional accredited investors

Can apply IFRS for SMEs which are less complex in numerous ways: § Topics not relevant for SMEs

are omitted; for example earnings per share, interim financial reporting and segment reporting

§ Many principles for recognising and measuring assets, liabilities, income and expenses in full IFRS Standards are simplified. For example, amortise goodwill; recognise all borrowing and development costs as expenses; cost model for associates and jointly-controlled entities; and undue cost or effort exemptions for specific requirements

§ Significantly fewer disclosures are required (roughly a 90 per cent reduction).

§ The Standard has been written in clear, easily translatable language

§ To further reduce the burden for SMEs, revisions are expected to be limited to once every three years

of 95

10.2 Appendix B – Table 7

Table 7 - Document Count Annu

al Re

port

Quar

terly

Rep

ort

Earn

ings

Call

Tra

nscr

ipt

Quar

terly

Pre

sent

ation

Earn

ings

Pre

ss R

eleas

eIn

vest

or P

rese

ntat

ion

Othe

r Sou

rces

Main Sample1 AO World plc x x x x2 Atlassian Corporation Plc x x x x x3 Boozt AB x x x x x4 Box, Inc. x x x x x5 Delivery Hero SE x x x x x6 HelloFresh SE x x x x7 LendingClub Corporation x x x x x8 LINE Corporation x x x x9 MyBucks S.A. x x x

10 Okta, Inc. x x x x x11 OnDeck Capital, Inc. x x x x x x12 ServiceNow, Inc. x x x x x13 Shop Apotheke Europe N.V. x x x x14 Shopify, Inc. x x x x x15 Snap Inc. x x x x x16 Splunk, Inc. x x x x x17 Square, Inc. x x x x x18 Tableau Software, Inc. x x x x19 Takeaway.com N.V. x x x x x20 Talend S.A. x x x x21 Twilio Inc. x x x x22 Twitter, Inc. x x x x x x23 windeln.de SE x x x x x24 Workday, Inc. x x x x25 Xero Limited x x x x xTotal Document Count 25 24 23 18 15 11 1

Control Sample1 Catena Media p.l.c x x x x x2 Grubhub Inc. x x x x3 Micro Focus International plc x x x x4 Netflix, Inc. x x x x x5 Paypal Holdings, Inc. x x x x x6 Rovio Entertainment Oyj x x x x7 salesforce.com, inc. x x x x x8 Scout24 AG x x x x x9 Veeva Systems Inc. x x x x x

10 Zalando SE x x x x xTotal Document Count 10 9 8 8 9 1 2Table 7 depicts the total document count for the 25 firms, which are part of the main sample, as well as for the 10 firms, which are part of the control sample. In total, 117 documents were read and codified for the main sample and 47 documents for the control sample. Other Sources contains further documents voluntarily provided by a firm within the Investor Relations section on the company webpage.

of 95

10.3 Appendix C – Node Definitions

Top

icSu

b-to

pic

Nod

eD

efin

ition

Sour

ceEx

empl

ary

Quo

te

Cost

of D

ebt

The r

ate o

f ret

urn

that

the l

ende

r req

uire

s, so

met

imes

ca

lled

the c

ost o

f cap

ital f

or d

ebt.

Penm

an (2

013)

“Whi

le in

tere

st ra

tes

have

rise

n an

d th

e fe

dera

l tax

rat

e is

now

lowe

r (r

educ

ing

the

tax

shie

ld o

n in

tere

st co

sts),

we ju

dge

that

our

afte

r-ta

x co

st of

deb

t con

tinue

s to

be

lowe

r th

an o

ur c

ost o

f equ

ity, s

o we

an

ticip

ate

that

we’

ll co

ntin

ue to

fina

nce

our

capi

tal n

eeds

in th

e hi

gh y

ield

mar

ket.”

Net

flix,

Let

ter

to

Shar

ehol

ders

Leve

rage

Ra

tio

Mea

sure

use

d in

the a

cade

mic

liter

atur

e defi

ned

as

mar

ket v

alue

of d

ebt (

D) o

ver t

he m

arke

t val

ue o

f deb

t pl

us eq

uity

(E).

This

ratio

mea

sure

s how

muc

h of

the

com

pany

’s en

terp

rise v

alue

is cl

aim

ed b

y de

bt h

olde

rs

Kol

ler et

al.

(201

6)“T

he c

apita

l stru

ctur

e as

of t

he b

alan

ce s

heet

dat

e is

char

acte

rized

by

an e

quity

rat

io o

f 67.

0% (

prev

ious

ye

ar 1

3.6%

). Th

e G

roup

is fi

nanc

ed b

y a

mix

ture

of e

quity

issu

ed th

roug

h th

e su

cces

sful I

PO a

nd p

rivat

e eq

uity

rou

nds

and

debt

fina

ncin

g:”

Hello

Fres

h, A

nnua

l Rep

ort

Free

Cas

h Fl

owTh

e cas

h flo

w ge

nera

ted

by th

e cor

e ope

ratio

ns o

f the

bu

sines

s afte

r ded

uctin

g in

vest

men

ts in

new

capi

tal

Kol

ler et

al.

(201

6)“F

ree

cash

flow

is e

xpec

ted

to b

e in

the

rang

e of

$35

0 m

illio

n to

$36

0 m

illio

n, w

hich

incl

udes

cap

ital

expe

nditu

res

that

are

exp

ecte

d to

be

appr

oxim

atel

y $4

0 m

illio

n in

fisc

al 2

019.

” At

lass

ian,

Pre

ss R

elea

se

NOPL

AT: n

et o

pera

ting

prof

its le

ss a

djus

ted

taxe

s

Non-

cash

ope

ratin

g ex

pens

es su

ch a

s dep

recia

tion

and

amor

tizat

ion.

Chan

ge in

ope

ratin

g wo

rkin

g ca

pita

l

Net c

apita

l exp

endi

ture

s suc

h as

inve

stm

ents

in

prop

erty

, pla

nt, a

nd eq

uipm

ent l

ess t

he b

ook

valu

e of

any

PP&

E so

ld

Chan

ge in

capi

talis

ed o

pera

ting

lease

s

Inve

stm

ent i

n go

odwi

ll an

d ac

quire

d in

tang

ibles

Chan

ge in

oth

er lo

ng-te

rm o

pera

ting

asse

ts, n

et o

f lo

ng-te

rm li

abili

ties

Term

inal

Va

lue G

rowt

h Ra

te

The r

ate a

t whi

ch th

e com

pany

’s NO

PLAT

and

cash

flo

w gr

ow ea

ch y

ear i

n st

eady

stat

eK

oller

et a

l. (2

016)

“Lon

g-te

rm g

rowt

h ra

te: T

he lo

ng te

rm g

rowt

h ra

te is

a r

efle

ctio

n of

the

loca

l Aus

tralia

n in

flatio

n ra

te fo

r th

e fiv

e ye

ar fo

reca

st pe

riod.

” M

yBuc

ks, A

nnua

l Rep

ort

WAC

CTh

e rat

e of r

etur

n th

at in

vest

ors e

xpec

t to

earn

from

in

vest

ing

in th

e com

pany

and

ther

efore

the a

ppro

pria

te

disc

ount

rate

for t

he fr

ee ca

sh fl

owK

oller

et a

l. (2

016)

“spe

cific

risk

pre

miu

m b

etwe

en 6

.0%

and

25.

0% (

prev

ious

yea

r: 6.

5% to

18.

0%).”

Del

iver

y He

ro, A

nnua

l Re

port

Cost

of

Equi

ty

The a

mou

nt th

at a

n in

vest

or re

quire

s to

com

pens

ate

her f

or th

e tim

e val

ue o

f mon

ey ti

ed u

p in

the

inve

stm

ent a

nd fo

r tak

ing

on ri

sk in

the i

nves

tmen

tPe

nman

(201

3)"T

he [.

..] s

hare

has

bee

n in

clud

ed in

to th

e M

SCI

Smal

l Cap

Ger

man

y In

dex

since

31

May

201

8, w

hich

at

tract

s in

tern

atio

nal i

nter

est a

nd h

as in

crea

sed

liqui

dity

in th

e sh

are.

" Sh

op A

poth

eke

Euro

pe, Q

uarte

rly

Pres

enta

tion

Trun

catio

n Po

int i

n Ti

me

Horiz

on p

oint

in ti

me t

= T

, su

ch th

at th

e bus

ines

s go

odwi

ll ca

n be

expe

cted

to b

e neg

ligib

le at

this

poin

t in

tim

eSk

ogsv

ik (2

002)

“Now

, I a

gree

with

you

phi

loso

phic

ally

that

whe

n we

get

to s

tead

y sta

te th

en w

e’re

goi

ng to

see

the

norm

aliza

tion

of c

ash

flow

yiel

d an

d m

argi

n de

liver

y an

d al

l tho

se ty

pes

of e

lem

ents,

but

I th

ink

we h

ave

to a

chie

ve th

at s

tead

y sta

te a

nd I

’m n

ot d

ecla

ring

that

we’

re a

t it y

et.”

Spl

unk,

Ear

ning

s Ca

ll

Financial Information

“Cap

Ex th

is ye

ar w

ill b

e ve

ry m

eani

ngfu

lly b

ack

end

load

ed w

hich

is r

elat

ed to

the

timin

g of

our

pro

ject

s, wh

ich

are

all p

rogr

essin

g in

line

with

pla

n, b

ut w

ill b

e ve

ry c

once

ntra

ted

in th

e se

cond

hal

f of t

he y

ear

and

we c

ontin

ue to

exp

ect E

UR35

0 m

illio

n Ca

pEx

for

the

full

year

.” Z

alan

do, E

arni

ngs

Call

Overlapping Model Inputs

Discounted Cash-Flow Model

Unde

rlyin

g Fr

ee C

ash

Flow

Lin

e It

ems

Kol

ler et

al.

(201

6)

of 95

Top

icSu

b-to

pic

Nod

eD

efin

ition

Sour

ceEx

empl

ary

Quo

te

Retu

rn o

n Eq

uity

Book

retu

rn o

n ow

ners

’ equ

ity, a

ccru

ed in

per

iod

tSk

ogsv

ik (2

002)

"Ret

urn

on e

quity

, rol

ling

12 m

onth

s, of

25%

" Ca

tena

Med

ia, Q

uarte

rly R

epor

t

Acco

untin

g ne

t inc

ome,

accr

ued

in p

erio

d t

Book

val

ue o

f own

ers’

equi

ty, a

ccru

ed in

Per

iod

t-1

Gro

wth

Rate

Eq

uity

Boo

k Va

lue

Rate

of c

hang

e of b

ook

valu

e of o

wner

s’ eq

uity

, de

term

ined

ex d

ivid

end

and

inclu

ding

any

new

issu

e of

shar

e cap

ital a

t tim

e tSk

ogsv

ik (2

002)

“Div

iden

d po

licy

Rovi

o’s

long

-term

goa

l is

to d

istrib

ute

appr

oxim

atel

y 30

per

cent

of a

nnua

l net

res

ults

excl

udin

g ite

ms

affe

ctin

g co

mpa

rabi

lity

as d

ivid

end

and

equi

ty r

etur

ns.”

Rov

io, A

nnua

l Rep

ort

Acco

untin

g M

easu

rem

ent

Bias

Disc

repa

ncies

bet

ween

a ‘t

rue a

nd fa

ir’ m

atch

ing

of

com

pany

reve

nues

and

cost

s, an

d th

e act

ual m

atch

ing

that

take

s pla

ce in

the a

ccou

ntin

g re

port

sSk

ogsv

ik (2

002)

“App

roxi

mat

ely

half

of th

e in

vestm

ent w

ill im

pact

Gro

up p

rofit

, i.e

. an

impa

ct o

f 2–3

%-p

oint

s to

the

expe

cted

EBI

T m

argi

n, a

nd h

alf w

ill b

e ca

pita

lised

dev

elop

men

t cos

ts an

d ad

vanc

e pa

ymen

ts.”

Rovi

o,

Annu

al R

epor

t

Supp

lemen

tal

Reve

nue

Info

rmat

ion

Addi

tiona

l non

-requ

ired

info

rmat

ion

abou

t rev

enue

s (e

.g. C

onst

ant c

urre

ncy

reve

nues

, non

-GAA

P re

venu

es,

reve

nues

fore

cast

s, re

venu

es b

reak

down

s)-

“We

now

expe

ct to

rea

ch g

rowt

h le

vels

at th

e up

per

end

of th

e ra

nge

of th

e fu

ll ye

ar g

uida

nce

we g

ave

to

the

mar

ket a

t the

beg

inni

ng o

f the

yea

r in

term

s of

rev

enue

gro

wth.

” Sc

out2

4, P

ress

Rel

ease

Supp

lemen

tal

Expe

nse

Info

rmat

ion

Addi

tiona

l non

-requ

ired

info

rmat

ion

abou

t exp

ense

s (e

.g. C

onst

ant c

urre

ncy

expe

nses

, non

-GAA

P ex

pens

es,

expe

nses

fore

cast

s, ex

pens

es b

reak

down

s)-

“Cos

t man

agem

ent p

rogr

ams

prog

ress

ing

well

with

the

Gro

up ta

rget

ing

furth

er $

300m

of a

nnua

lized

cos

t sa

ving

s by

end

of F

Y20.

” M

icro

Foc

us, P

ress

Rel

ease

Alte

rnat

ive

Acco

untin

g M

etric

s

Expl

icitly

men

tione

d no

n-G

AAP

prof

itabi

lity

mea

sure

s or r

atio

s-

“We

expe

ct th

ird q

uarte

r ad

juste

d EB

ITDA

to b

e in

the

rang

e of

$58

mill

ion

to $

64 m

illio

n an

d ar

e in

crea

sing

our

full

year

EBI

TDA

guid

ance

to a

ran

ge o

f $25

6 m

illio

n to

$27

0 m

illio

n. T

his

incl

udes

som

e in

crem

enta

l mar

ketin

g in

vestm

ent i

n th

e th

ird q

uarte

r, de

signe

d to

cap

italiz

e on

the

stron

g di

ner

grow

th

trend

s we

saw

in th

e se

cond

qua

rter

that

hav

e ex

tend

ed in

to th

e be

ginn

ing

of th

e th

ird q

uarte

r, as

wel

l as

the

impa

ct o

f our

pre

viou

sly d

iscus

sed

inve

stmen

t in

deliv

ery

expa

nsio

n.”

Gru

bhub

, Ear

ning

s Ca

ll

Busin

ess

Mod

el Sp

ecifi

c K

PIs

Expl

icitly

men

tione

d co

mpa

ny sp

ecifi

c key

per

form

ance

in

dica

tors

-

“Ave

rage

bas

ket s

ize: s

imila

r to

the

num

ber

of o

rder

s pl

aced

, the

ave

rage

bas

ket s

ize h

as a

dire

ct e

ffect

on

the

reve

nue

of th

e gr

oup.

The

ave

rage

bas

ket s

ize (

afte

r re

turn

s) d

ecre

ased

slig

htly

in fi

scal

yea

r 20

17

from

EUR

66.

6 to

EUR

64.

5. I

t is

influ

ence

d by

ass

ortm

ent c

ompo

sitio

n, c

usto

mer

age

, and

sho

ppin

g ch

anne

l. Yo

ung

custo

mer

s wh

o pr

efer

fast

fash

ion

artic

les

and

shop

ping

mob

ile te

nd to

sho

p m

ore

frequ

ently

, but

with

a lo

wer

bask

et s

ize.”

Zal

ando

, Ann

ual R

epor

t

“We

are

now

targ

etin

g no

n-G

AAP

net i

ncom

e pe

r sh

are

of b

etwe

en $

1.47

and

$1.

48, b

ased

on

a fu

lly

dilu

ted

shar

e co

unt o

f app

roxi

mat

ely

$156

mill

ion.

” Ve

eva,

Ear

ning

s Ca

llAdditional Financial Data

Residual Income Valuation Model

Unde

rlyin

g Re

turn

on

Equi

ty It

ems

Skog

svik

(200

2)

Financial Information (continued)

of 95

Top

icSu

b-to

pic

Nod

eD

efin

ition

Sour

ceEx

empl

ary

Quo

te

Stra

tegy

Ex

ecut

ion

&

Cons

isten

cy

Conc

rete

pla

ns o

r act

ions

that

dire

ctly

serv

e the

pu

rpos

e of f

ulfil

ling

the c

ompa

ny´s

gro

wth

stra

tegy

Hoffm

ann

& F

iesele

r (2

012)

“Our

prio

rity

next

yea

r wi

ll be

to le

vera

ge o

ur e

xisti

ng c

ompe

tenc

ies

whils

t exp

lorin

g ot

hers

whe

re w

e ca

n ap

ply

our

skill

s an

d kn

owle

dge

and

incr

ease

pro

fitab

ility

. For

exa

mpl

e, w

e wi

ll fo

cus

mor

e on

our

th

ird p

arty

logi

stics

and

rec

yclin

g bu

sines

ses,

whic

h we

exp

ect t

o in

vest

in fu

rther

ove

r th

e co

min

g ye

ars.”

AO

Wor

ld, A

nnua

l Rep

ort

Gro

wth

Stra

tegy

Dire

ctio

n in

the f

irm’s

mar

ket p

ositi

onin

g,

inte

ract

ions

acr

oss o

rgan

izatio

nal b

ound

aries

Mor

ris et

al.

(200

5)

“Key

ele

men

ts of

our

stra

tegy

incl

ude:

cro

ss s

ell a

nd u

psel

l; ex

tend

exi

sting

ser

vice

offe

rings

; red

uce

custo

mer

attr

ition

; exp

and

and

stren

gthe

n th

e pa

rtner

eco

syste

m; e

xpan

d in

tern

atio

nally

; tar

get v

ertic

al

indu

strie

s; ex

pand

into

new

hor

izont

al m

arke

ts; e

xten

d go

-to-m

arke

t cap

abili

ties;”

Sal

esfo

rce,

Qua

rterly

Re

port

Valu

e Pr

opos

ition

The n

atur

e of t

he p

rodu

ct/

serv

ice m

ix, t

he fi

rm’s

role

in p

rodu

ctio

n or

serv

ice d

elive

ry, a

nd h

ow th

e offe

ring

is m

ade a

vaila

ble t

o cu

stom

ers;

salie

nt p

oint

s of

diffe

renc

e tha

t can

be m

aint

aine

d

Mor

ris et

al.

(200

5)“O

ur k

ey o

fferin

g in

our

cor

e re

tail

busin

ess

rem

ains

stro

ng; u

nbea

tabl

e pr

ices

, hug

e ra

nge,

wid

e av

aila

bilit

y, s

mar

t and

inno

vativ

e we

b co

nten

t and

am

azin

g se

rvic

e.”

AO W

orld

, Ann

ual R

epor

t

Miss

ion,

Vi

sion

&

Valu

es

Resp

ectiv

ely, a

com

pany

’s pu

rpos

e, its

asp

iratio

n fo

r fu

ture

resu

lts, a

nd th

e int

erna

l com

pass

that

will

gui

de

its a

ctio

nsK

apla

n &

Nor

ton

(200

8)“A

t Box

, our

miss

ion

is to

pow

er h

ow th

e wo

rld w

orks

toge

ther

.” B

ox, A

nnua

l Rep

ort

M&

A St

rate

gyEx

plici

t str

ateg

ies, p

lans

or t

arge

ts fo

r mer

gers

and

/ or

acq

uisit

ions

defi

ned

by th

e com

pany

-“M

yBuc

ks to

ok a

del

iber

ate

deci

sion

to a

cqui

re h

eavi

ly d

istre

ssed

pla

tform

s, wh

ich

it co

uld

turn

aro

und

with

its

oper

atio

nal s

kills

, to

min

imize

the

effe

ctiv

e ca

pita

l cos

ts of

acq

uirin

g th

e Ba

nks.”

MyB

ucks

, An

nual

Rep

ort

Cust

omer

In

form

atio

nCu

stom

er ty

pes,

their

geo

grap

hic d

isper

sion,

and

their

in

tera

ctio

nM

orris

et a

l. (2

005)

“Gre

w pa

ying

cus

tom

er b

ase

to m

ore

than

87,

000

busin

esse

s, in

clud

ing

new

or e

xpan

ded

depl

oym

ents

with

le

adin

g or

gani

zatio

ns s

uch

as C

anon

U.S

.A.,

City

of A

tlant

a, J

LL, L

ions

gate

, Nat

ionw

ide,

The

Ph

ilade

lphi

a Ph

illie

s, Ro

dan

& F

ield

s, Sa

cram

ento

Kin

gs, S

ocie

te G

ener

ale

U.S.

and

Wor

ld F

uel

Serv

ices

.” B

ox, P

ress

Rel

ease

Empl

oyee

In

form

atio

nAd

ditio

nal i

nfor

mat

ion

rela

ted

to em

ploy

ees,

key

pers

onne

l, em

ploy

ee in

itiat

ives

-

“The

re is

inte

nse

com

petit

ion

for

high

-end

tale

nt in

the

gam

es in

dustr

y, a

nd w

e ar

e al

ways

look

ing

for

com

pete

nt p

rofe

ssio

nals

in F

inla

nd a

nd a

broa

d. T

he le

ader

s of

our

gam

e de

velo

pmen

t stu

dios

cul

tivat

e th

e cr

eativ

e an

d te

chni

cal e

xper

tise

of o

ur e

mpl

oyee

s in

var

ious

way

s; fo

r ex

ampl

e, b

y gi

ving

them

op

portu

nitie

s to

dev

elop

thei

r sk

ills

and

com

e up

with

new

insig

hts

at w

ork.

It i

s im

porta

nt to

mak

e a

cons

ciou

s ef

fort

to p

rovi

de s

pace

for

crea

tivity

and

pas

sion

to e

nabl

e in

nova

tive

prod

ucts.

” Ro

vio,

Ann

ual

Repo

rt

R&D

and

Proj

ect

Activ

ity

Expl

icit p

lans

and

act

iviti

es fo

r res

earc

h an

d de

velo

pmen

t pro

jects

-“W

e ha

ve s

ough

t to

rapi

dly

impr

ove

the

capa

bilit

ies

of o

ur p

rodu

cts

over

tim

e an

d in

tend

to c

ontin

ue to

in

vest

in p

rodu

ct in

nova

tion

and

lead

ersh

ip.”

Tab

leau

, Ann

ual R

epor

t

Business Model

Operating Information

of 95

Top

icSu

b-to

pic

Nod

eD

efin

ition

Sour

ceEx

empl

ary

Quo

te

Barg

aini

ng

Powe

r of

Supp

liers

Powe

rful s

uppl

iers c

aptu

re m

ore o

f the

val

ue fo

r th

emse

lves

by

char

ging

hig

her p

rices

, lim

iting

qua

lity

or se

rvice

s, or

shift

ing

cost

s to

indu

stry

par

ticip

ants

Port

er (2

008)

“Our

abi

lity

to p

rovi

de o

ur m

embe

rs w

ith c

onte

nt th

ey c

an w

atch

dep

ends

on

studi

os, c

onte

nt p

rovi

ders

an

d ot

her

right

s ho

lder

s lic

ensin

g rig

hts

to d

istrib

ute

such

con

tent

and

cer

tain

rel

ated

ele

men

ts th

ereo

f, su

ch a

s th

e pu

blic

per

form

ance

of m

usic

con

tain

ed w

ithin

the

cont

ent w

e di

strib

ute.

The

lice

nse

perio

ds

and

the

term

s an

d co

nditi

ons

of s

uch

licen

ses

vary

. If t

he s

tudi

os, c

onte

nt p

rovi

ders

and

oth

er r

ight

s ho

lder

s ar

e no

t or

are

no lo

nger

will

ing

or a

ble

to li

cens

e us

con

tent

upo

n te

rms

acce

ptab

le to

us,

our

abili

ty to

stre

am c

onte

nt to

our

mem

bers

will

be

adve

rsel

y af

fect

ed a

nd/o

r ou

r co

sts c

ould

incr

ease

. Ce

rtain

lice

nses

for

cont

ent p

rovi

de fo

r th

e stu

dios

or

othe

r co

nten

t pro

vide

rs to

with

draw

con

tent

from

ou

r se

rvic

e re

lativ

ely

quic

kly.

” Ne

tflix

, Ann

ual R

epor

t

Barg

aini

ng

Powe

r of

Cust

omer

s

Powe

rful c

usto

mer

s can

capt

ure m

ore v

alue

by

forc

ing

down

pric

es, d

eman

ding

bet

ter q

ualit

y or

mor

e ser

vice

(t

here

by d

rivin

g up

cost

s), a

nd g

ener

ally

pla

ying

in

dust

ry p

artic

ipan

ts o

ff ag

ains

t one

ano

ther

, all

at

the e

xpen

se o

f ind

ustr

y pr

ofita

bilit

y

Port

er (2

008)

“The

loss

of o

ne o

r m

ore

of o

ur k

ey c

usto

mer

s, or

a fa

ilure

to r

enew

our

sub

scrip

tion

agre

emen

ts wi

th

one

or m

ore

of o

ur k

ey c

usto

mer

s, co

uld

nega

tivel

y af

fect

our

abi

lity

to m

arke

t our

app

licat

ions

” W

orkd

ay,

Qua

rtely

Rep

ort

Thre

at o

f Su

bstit

utes

A su

bstit

ute p

erfo

rms t

he sa

me o

r a si

mila

r fun

ctio

n as

an

indu

stry

’s pr

oduc

t by

a di

ffere

nt m

eans

Port

er (2

008)

“Add

ition

ally

, we

face

com

petit

ion

from

the

offli

ne g

roce

ry r

etai

l and

onl

ine/

offli

ne g

roce

ry d

eliv

ery

serv

ice

prov

ider

s” H

ello

Fres

h, A

nnua

l Rep

ort

Thre

at o

f New

En

tran

ts

The t

hrea

t of e

ntry

in a

n in

dust

ry d

epen

ds o

n th

e he

ight

of e

ntry

bar

riers

that

are

pre

sent

and

on

the

reac

tion

entr

ants

can

expe

ct fr

om in

cum

bent

sPo

rter

(200

8)

“Bec

ause

of t

he c

hara

cter

istic

s of

ope

n so

urce

sof

twar

e, th

ere

are

few

tech

nolo

gica

l bar

riers

to e

ntry

into

th

e op

en s

ourc

e m

arke

t by

new

com

petit

ors

and

it m

ay b

e re

lativ

ely

easy

for

com

petit

ors,

som

e of

who

m

may

hav

e gr

eate

r re

sour

ces

than

we

have

, to

ente

r ou

r m

arke

ts an

d co

mpe

te w

ith u

s. O

ne o

f the

ch

arac

teris

tics

of o

pen

sour

ce s

oftw

are

is th

at a

nyon

e m

ay o

btai

n ac

cess

to th

e so

urce

cod

e fo

r ou

r op

en

sour

ce p

rodu

cts

and

then

mod

ify a

nd r

edist

ribut

e th

e ex

istin

g op

en s

ourc

e so

ftwar

e an

d us

e it

to c

ompe

te

in th

e m

arke

tpla

ce.”

Tal

end,

Ann

ual R

epor

t

Indu

stry

Ri

valry

Riva

lry a

mon

g ex

istin

g co

mpe

titor

s tak

es m

any

fam

iliar

form

s, in

cludi

ng p

rice d

iscou

ntin

g, n

ew

prod

uct i

ntro

duct

ions

, adv

ertis

ing

cam

paig

ns, a

nd

serv

ice im

prov

emen

ts. H

igh

rival

ry li

mits

the

prof

itabi

lity

of a

n in

dust

ry

Port

er (2

008)

“We

prim

arily

com

pete

with

the

tradi

tiona

l offl

ine

orde

ring

proc

ess

used

by

the

vast

maj

ority

of

resta

uran

ts an

d di

ners

invo

lvin

g th

e te

leph

one

and

pape

r m

enus

that

res

taur

ants

distr

ibut

e to

din

ers,

as

well

as a

dver

tisin

g th

at r

esta

uran

ts pl

ace

in lo

cal p

ublic

atio

ns to

attr

act d

iner

s.” G

rubh

ub, A

nnua

l Rep

ort

Mar

ket

Cond

ition

Past

, cur

rent

and

futu

re st

ate o

f the

mar

ket(

s) a

nd th

e ec

onom

y(ies

) in

which

the c

ompa

ny is

act

ive

-“A

nd w

ith a

n on

line

mar

ket g

rowt

h of

10-

15%

in th

e No

rdic

s an

d a

cont

inue

d co

nsol

idat

ion

wher

e th

e sm

alle

r pl

ayer

s ar

e lo

sing

shar

e to

the

larg

er p

laye

rs th

e gr

owth

pro

spec

ts ar

e br

ight

and

we

aim

to ta

ke

mor

e th

an o

ur “

fair

shar

e of

the

mar

ket”

Boo

zt, A

nnur

al R

epor

t

Mar

ket S

hare

Perc

enta

ge o

f mar

ket a

ccou

nted

for b

y a

spec

ific

com

pany

; key

indi

cato

r of c

ompe

titiv

enes

sFa

rris

et a

l. (2

010)

“Bei

ng a

lead

er in

use

r tra

ffic

and

enga

gem

ent,

we a

re w

ell-p

ositi

oned

to b

enef

it fro

m th

e re

venu

e an

d gr

owth

pot

entia

ls in

the

larg

e ad

jace

nt m

arke

t seg

men

ts ou

tside

our

cor

e cl

assif

ieds

bus

ines

s, be

it th

e va

lue

chai

n fo

r th

e en

tire

prop

erty

pur

chas

e or

ren

tal p

roce

ss, o

r fo

r th

e au

tom

otiv

e m

arke

t.” S

cout

24,

Annu

al R

epor

t

Inno

vatio

n Tr

ends

Curr

ent a

nd fu

ture

dev

elopm

ent i

n te

chno

logy

(ies)

th

at co

uld

be d

isrup

tive f

or th

e com

pany

-

“No

disr

uptiv

e te

chno

logy

has

bee

n in

trodu

ced

into

our

mar

kets

and

the

grow

th in

our

net

work

of b

oth

cons

umer

s an

d re

staur

ants

dem

onstr

ates

the

cont

inui

ng a

nd in

crea

sing

stren

gth

of o

ur v

alue

pro

posit

ion.

W

e ha

ve a

lso c

ontin

ued

to in

vest

in th

e fu

nctio

nalit

y of

our

pro

duct

to m

ake

the

cons

umer

exp

erie

nce

as

smoo

th a

s po

ssib

le.”

Tak

eawa

y, A

nnua

l Rep

ort

Regu

lato

ry

Fram

ewor

kPa

st, c

urre

nt a

nd fu

ture

stat

e of t

he la

ws a

nd

regu

latio

ns th

at a

com

pany

has

to co

mpl

y wi

th-

“Sev

eral

sta

tes

are

curr

ently

pro

posin

g va

rious

form

s of

reg

ulat

ion

targ

etin

g on

line

gam

ing,

incl

udin

g a

rece

nt o

nlin

e po

ker

bill

in N

ew Y

ork

that

is li

kely

to b

e pa

ssed

by

the

state

sen

ate.

Fur

ther

mor

e, th

e Su

prem

e Co

urt i

s ta

king

up

a ca

se th

at c

ould

mak

e sp

orts

betti

ng w

idel

y av

aila

ble

in th

e US

. Cat

ena

Med

ia is

wel

l-pos

ition

ed to

cap

italis

e on

thes

e po

sitiv

e op

portu

nitie

s.” C

aten

a M

edia

, Ann

ual R

epor

t

Operating Information (continued)

Competitive Landscape Market Environment

of 95

Top

icSu

b-to

pic

Nod

eD

efin

ition

Sour

ceEx

empl

ary

Quo

te

Bran

d St

reng

thTh

e pop

ular

ity a

nd p

ublic

ity o

f a co

mpa

ny’s

bran

d(s)

Hoffm

ann

& F

iesele

r (2

012)

“The

Boo

zt br

and

is be

com

ing

a re

cogn

ised

nam

e fo

r fa

shio

n in

the

Nord

ics

thro

ugh

high

cus

tom

er

satis

fact

ion.

Thi

s is

cons

isten

tly p

rove

n by

an

NPS

of 6

5, a

Tru

stpilo

t sco

re o

f 9.2

(De

cem

ber

2017

), an

d a

grow

ing

base

of r

etur

ning

cus

tom

ers.”

Boo

zt, A

nnua

l Rep

ort

Publ

ic Re

puta

tion

The r

eput

atio

n of

a co

mpa

ny in

the p

ublic

sphe

reHo

ffman

n &

Fies

eler

(201

2)

“Ear

lier

this

mon

th, K

eith

Wee

d, th

e Ch

ief M

arke

ting

Offi

cer

of U

nile

ver,

the

seco

nd la

rges

t adv

ertis

er in

th

e wo

rld, p

ublic

ly p

raise

d ou

r ef

forts

. Kei

th, w

ho is

nam

ed b

y Fo

rbes

as

one

of th

e wo

rld’s

mos

t in

fluen

tial C

MO

s, tw

eete

d, "

Plea

sed

to s

ee T

witte

r ta

king

a b

ig s

tand

aga

inst

the

fake

follo

wers

pol

lutin

g th

e di

gita

l eco

syste

m. G

reat

ste

p fo

rwar

d wh

ich

stren

gthe

ns th

e in

dustr

y. I

hop

e to

see

mor

e fo

llowi

ng."”

Tw

itter

, Ear

ning

s Ca

ll

Degr

ee o

f Co

mm

itmen

tLe

vel o

f pre

cisio

n in

fore

cast

gui

danc

e pro

vide

d by

the

com

pany

, i.e.

vag

ue, d

irect

iona

l, nu

mer

ical t

arge

tG

raha

m et

al.

(200

5)

“On

aver

age,

we

are

targ

etin

g $1

bill

ion

to $

3 bi

llion

of M

&A

per

year

and

ret

urni

ng 4

0% to

50%

of o

ur

free

cash

flow

to s

hare

hold

ers

over

that

per

iod.

We

are

also

com

mitt

ed to

opt

imizi

ng o

ur c

apita

l stru

ctur

e,

while

mai

ntai

ning

an

inve

stmen

t gra

de r

atin

g, b

alan

cing

inor

gani

c an

d or

gani

c in

vesti

ng w

ith m

argi

n ex

pans

ion

and

cont

inui

ng o

ur s

trate

gy to

div

ersif

y fu

ndin

g so

urce

s fo

r ou

r cr

edit

busin

ess

to r

educ

e ca

pita

l in

tens

ity.”

Pay

Pal,

Earn

ings

Cal

l

Exte

rnal

So

urce

Re

lianc

e

Expl

icit m

entio

ning

of e

xter

nal s

ourc

es u

sed

for

inte

rnal

calcu

latio

n, es

timat

es o

r ass

umpt

ions

Gra

ham

et a

l. (2

005)

“In

Mar

ch 2

018,

the

IMF

issue

d in

its

Wor

ld E

cono

mic

Out

look

a g

loba

l eco

nom

ic g

rowt

h of

3.9

% fo

r 20

18. T

his

grow

th is

exp

ecte

d to

be

driv

en b

y a

num

ber

of fa

ctor

s, su

ch a

fisc

al s

timul

us a

nd in

vestm

ents.

Ho

weve

r, th

e IM

F al

so s

ees

a co

uple

of r

isk fa

ctor

s th

at c

an d

erai

l the

rec

over

y –

mai

nly

refe

rrin

g to

po

litic

al a

nd tr

ade

tens

ions

as

well

as h

igh

debt

.” D

eliv

ery

Hero

, Ann

ual R

epor

t

Hist

orica

l Fo

reca

st

Accu

racy

Expl

icit m

entio

ning

of w

heth

er p

revi

ous g

uida

nce

prov

ided

in ea

rlier

repo

rts w

as m

et b

y th

e com

pany

Gra

ham

et a

l. (2

005)

“Tre

llo r

even

ue, w

hich

is in

clud

ed in

sub

scrip

tion

reve

nue,

mea

ning

fully

exc

eede

d ou

r ta

rget

of

appr

oxim

atel

y $2

0 m

illio

n fo

r fis

cal 2

018

that

we

had

prev

ious

ly s

et o

ut a

t the

beg

inni

ng o

f the

yea

r.”

Atla

ssia

n, S

hare

hold

er L

ette

r

Credibility

General Reputation

Fact Check

of 95

10.4 Appendix D – Table 8

Tab

le 8

- R

ecor

ded

The

me

Dist

ribut

ion

in A

ccou

ntin

g St

anda

rd C

lust

ers

M

ain

Sam

ple

Tot

alC

ount

Disc

ount

ed

Cash

-Flo

wO

verla

ppin

g M

odel

Inpu

tsR

esid

ual I

ncom

e V

alua

tion

Add

ition

al

Fina

ncia

l Dat

aBu

sines

sM

odel

Com

petit

ive

Land

scap

eM

arke

t En

viro

nmen

tG

ener

al

Rep

utat

ion

IFR

STo

tal N

umbe

r of

Rec

orde

d Th

emes

1,64

211

23

4757

254

360

243

62%

of T

otal

6.8%

0.2%

2.9%

34.8

%33

.1%

3.7%

14.8

%3.

8%N

umbe

r of

Rec

orde

d Th

emes

62

44

85

42

Mea

n R

ecor

ded

Them

es p

er U

nder

lyin

g N

ode

18.7

1.5

11.8

143.

067

.912

.060

.831

.0N

umbe

r of

Sam

ple

Firm

s12

U.S

.-G

AA

PTo

tal N

umbe

r of

Rec

orde

d Th

emes

1,95

511

51

132

763

596

146

157

45%

of T

otal

5.9%

0.1%

6.8%

39.0

%30

.5%

7.5%

8.0%

2.3%

Num

ber

of R

ecor

ded

Them

es6

24

48

54

2M

ean

Rec

orde

d Th

emes

per

Und

erly

ing

Nod

e19

.20.

533

.019

0.8

74.5

29.2

39.3

22.5

Num

ber

of S

ampl

e Fi

rms

13

C

ontr

ol S

ampl

eT

otal

Cou

ntD

iscou

nted

Ca

sh-F

low

Ove

rlapp

ing

Mod

el In

puts

Res

idua

l Inc

ome

Val

uatio

nA

dditi

onal

Fi

nanc

ial D

ata

Busin

ess

Mod

elCo

mpe

titiv

e La

ndsc

ape

Mar

ket

Envi

ronm

ent

Gen

eral

R

eput

atio

nIF

RS

Tota

l Num

ber

of R

ecor

ded

Them

es87

912

10

5132

723

311

111

25%

of T

otal

13.8

%-

5.8%

37.2

%26

.5%

1.3%

12.6

%2.

8%N

umbe

r of

Rec

orde

d Th

emes

62

44

85

42

Mea

n R

ecor

ded

Them

es p

er U

nder

lyin

g N

ode

20.2

-12

.881

.829

.12.

227

.812

.5N

umbe

r of

Sam

ple

Firm

s5

U.S

.-G

AA

PTo

tal N

umbe

r of

Rec

orde

d Th

emes

695

360

5524

722

151

7411

% o

f Tot

al5.

2%-

7.9%

35.5

%31

.8%

7.3%

10.6

%1.

6%N

umbe

r of

Rec

orde

d Th

emes

62

44

85

42

Mea

n R

ecor

ded

Them

es p

er U

nder

lyin

g N

ode

6.0

-13

.861

.827

.610

.218

.55.

5N

umbe

r of

Sam

ple

Firm

s5

Tab

le 8

pro

vide

s des

crip

tive s

tatis

tics o

f the

tota

l rec

orde

d th

emes

acr

oss 2

5 fir

ms w

hich

are

par

t of t

he m

ain

sam

ple a

nd 1

0 fir

ms w

hich

are

par

t of t

he co

ntro

l sam

ple.

The r

espe

ctiv

e sam

ples

are

clus

tere

d ac

cord

ing

to th

e Acc

ount

ing

Stan

dard

s the

und

erly

ing

com

pani

es a

re re

port

ing

unde

r, i.e

. IFR

S a

nd U

.S.-G

AAP

. A re

cord

ed th

eme c

an ra

nge f

rom

a si

ngle

word

to a

sing

le do

cum

ent s

ectio

n as

a p

oten

tial r

ecor

ding

uni

t. Th

e cod

ifica

tion

mat

rix a

pplie

d is

sepa

rate

d in

to th

ree t

opics

: (1)

Fin

anci

al I

nfor

mat

ion

, (2)

O

pera

ting

Info

rmat

ion

and

(3) C

redi

bilit

y wi

th D

iscou

nted

Cas

h-Fl

ow M

odel

, Ove

rlapp

ing

Mod

el I

nput

s, R

esid

ual I

ncom

e Va

luat

ion

Mod

el a

nd A

dditi

onal

Fin

anci

al D

ata

as su

b-to

pics

belo

ngin

g to

(1),

Busin

ess

Mod

el, C

ompe

titiv

e La

ndsc

ape

and

Mar

ket E

nviro

nmen

t as

sub-

topi

cs b

elong

ing

to (2

) and

Gen

eral

Rep

utat

ion

as s

ub-to

pic b

elong

ing

to (3

). M

ean

Reco

rded

The

mes

per

Und

erly

ing

Node

is in

trod

uced

as a

mea

sure

of c

ompa

rabi

lity

due t

o di

ffere

nces

in th

e num

ber o

f nod

es u

nder

lyin

g a

sub-

topi

c.

of 95

10.5 Appendix E – Figure 8 & 9

Figure 8 Qualitative analysis summary for Forecast Issue 1 (control sample) using Gioia Methodology


§ Organic & inorganic growth on the agenda§ Clearly formulated investment criteria &

M&A strategy; disclose track record§ High strategic consistency across time§ Plan to strongly grow customer base (KPIs)§ Mention examples of strategic measures§ Marginal changes in growth & R&D strategy§ Overall little geographic expansion planned


(g) Current Competitive Edge

(i) Strategic Initiatives

§ Complexity reduction & ease of use§ Aim to ‘bridge old and new world’ § Service-/ Product-specific competitive edge§ Multi-faceted competitive edge§ Use ‘value proposition’ & ‘investment case’

(1) Excess ProfitabilityGeneration

(a*) Profit Generation

(c) Financial Myopia

(d) Business Roll-out

(e) Key Stakeholder Management

(f) ‘Favourable’ Business Environment

§ Low entry-barriers due to open software offer§ Fragmented markets which are about to

consolidate; traditional competitors exist§ In general large & dynamic markets§ Favourable macro-economic conditions§ ‘First-mover advantage’ in the market

§ Mention total customers & high dependence on subscription base renewals

§ Employees as ‘key asset’; plan to hire§ Competitive job market; need for training§ High dependence on key personnel§ Generally high bargaining power of suppliers

with 1 -2 key suppliers; no mitigation efforts

§ Break-down growth strategy in phases§ Mention short-term strategy execution§ Geographical expansion to a lesser extent§ Integrate innovation trends in offering§ Justify business model for growth strategy

§ Short-term targets in Business Models KPIs§ Mid-term target in Acc. Metrics§ Short-term revenue guidance, i.e. year-end§ Short-term EPS guidance, i.e. year-end

§ No book return on owners’ equity guidance§ Heavy use of EBIT(DA) & profitability§ Free Cash Flow guidance§ Guide on underlying Free Cash Flow items:

CAPEX, Working Capital, Cash Conversion§ Often raise revenue guidance§ Specific revenue guidance using growth rates§ Expense guidance/ targets on item level§ Expenses expected to grow; link to margins

of 95

Figure 9 Qualitative analysis summary for Forecast Issue 2 (control sample) using Gioia Methodology

1 st Order Concepts 2 nd Order Themes AggregateDimensions

(4) Internal Goodwill

Generation

(m) Internal Brand Strength

(n) External Brand Strength

§ Initiatives aimed to increase brand awareness§ Story-telling & anecdotes of recent successes

§ Actively mention recently won awards§ ‘Name-drop’ customers & outside partners § Customer relationship core part of mission§ Report customer success stories


(k) ‘True’ Operational Performance

(j) Supplemental Per-formance Reporting

§ Adjusted: Expenses, Net Income, Free Cash Flows, Acc. Metrics

§ Non-GAAP: Expenses, Net Income, EPS, Free Cash Flows, Acc. Metrics

§ Constant Currency: Revenues, Acc. Metrics§ Company-specific types of adjustments

§ Highlight their asset-light business models§ Capitalisation rate of intangible assets

(l) Asset Capitalisation

§ Leadership measured in individual KPIs§ KPIs link to revenue growth; little guidance

RIVolutionising Voluntary Disclosure

Documents

Transcript of RIVolutionising Voluntary Disclosure